(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Children's Library | Biodiversity Heritage Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "Nutritional ecology of Old-World fruit bats : a test of the calcium-constraint hypothesis"

NUTRITIONAL ECOLOGY OF OLD-WORLD FRUIT BATS: 
A TEST OF THE CALCIUM-CONSTRAINT HYPOTHESIS 



By 

SUZANNE LINN NELSON 



A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL 

OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT 

OF THE REQUIREMENTS FOR THE DEGREE OF 

DOCTOR OF PHILOSOPHY 

UNIVERSITY OF FLORIDA 

2003 



COPYRIGHT 2003 

by 
Suzanne Linn Nelson 



ACKNOWLEDGMENTS 

I am heavily indebted and very appreciative to my funding sources, without which 
this work could not have been completed. I appreciate the generous support of the Luis F. 
Bacardi Graduate Fellowship, the National Park of American Samoa, the Disney Wildlife 
Conservation Fund, and Bat Conservation International. I would also like to thank the 
members of my committee, Drs. Steve Humphrey (chair), Tom Kunz (co-chair), Mel 
Sunquist, George Tanner, and Lee McDowell for their support and advice during my 
studies and research. I appreciate their time, advice, and technical assistance on many 
aspects of this project. Their help was very appreciated. 

I thank Roger Haagenson and John Seyjagat (Lubee Foundation, Inc.) for 
offering and administering the Bacardi Graduate Fellowship. Brian Pope and Kerri Van 
Wormer were helpful in my understanding of the issues and care of captive flying foxes. 
Dana LeBlanc was always helpful answering questions concerning captive bat nutrition 
and management. Dr. Ellen Dierenfeld of the New York Zoological Society provided 
helpful nutrition discussions and important wildlife nutrition literature. 

There are many people to thank in American Samoa. I am grateful to the National 
Park of American Samoa for their funding of my work and for their generosity with 
trucks and field equipment. I extend special thanks to Charles Cranfield and Dr. Peter 
Craig for their advice, generosity, kindness, and support, which truly influenced the 
outcome of the research. I thank Epi Suafua and the staff of the National Park of 
American Samoa for their time, advice, and help with my project. 

iv 



I am very grateful to the U.S.D.A. Land Grant in American Samoa for building 
the bat house and providing equipment and supplies. I extend grateful thanks to the 
Forestry Crew, which includes Aitasi Sameli, Kitiona Fa'atamala, Ritofu Lotovale, 
Falaniko Mika, Tony Magalei, Logona Misa, and Eric Pese. I thank Dr. Carol Whitacker, 
Sheri Mann, and Orlo Collin Steele for their support of the bat house project and the land 
donation for its placement. I give special thanks to Dr. Don Vargo for his advice and help 
on many topics and for the use of his lab and drying ovens. 

I thank my field assistants, Raymond Pasay, Siniva Satele, and Gasetoto Gasetoto 
for their hard work. I especially thank Gasetoto Gasetoto for his exceptional efforts, 
including climbing and setting netting lines high in coconut trees. Joe Satele generously 
allowed us to use his land and house while we netted bats. I thank the Fuimono family at 
Fagatele Bay for their patience as we walked on their property at all hours. I thank Ian 
Gurr for showing me plots and bat roosts on the north side of the island. I also thank my 
friends in American Samoa whose company and help with netting I enjoyed very much, 
including Kelby Black, Alison Graves, Malcolm Gaylord, Catherine Buchanon, Josh 
Craig, Martin McCarthy, Graham Dawson, and Shintaro and Sophia Okamoto. I also 
thank Dan Smith and Jeff Marsh for the generous use of the hyperbaric chamber and their 
diving accident expertise. 

I wish to thank my friends and fellow Wildlife Ecology and Conservation 
graduate students for their help, support, and encouragement while at the University of 
Florida. I learned much from them and very much enjoyed talking with them about 
Biology. I wish to thank Caprice McRae, Monica Lindberg, Polly Falcon, Patty Connolly 
and Nat Frazer for always working on my behalf in financial matters and for providing 






support and encouragement. I thank Nancy Wilkinson and Jan Kivipelto of Animal 
Sciences for their help and advice on mineral analysis and laboratory techniques. 
Statistical help was generously provided by Dr. Ray Littell, Dr. Ken Portier, and 
Yongsung Joo. Dr. Clarence Ammerman provided insightful discussions on mineral 
bioavailability and apparent absorption. I thank Cindy Whitehurst of Park Avenue 
Women's Center for sharing information on bone densiometry and osteoporosis. Dr. 
Robert Barclay's ideas were the inspiration behind this dissertation. I thank him for his 
insight and support of my ideas and research. His essays and ideas on calcium were 
intriguing and inspiring. 

I thank my family and friends for their support and encouragement during my 
many years in school. Dr. Ed Heske and Dr. Lowell Getz always supported me and 
believed in an undergrad with lots of energy and bad grades. I am very grateful to have 
been visited in American Samoa by Doug Nelson and Laurie and Scott Beavers. I also 
thank Doug for providing pictures for my seminar. Both my immediate and extended 
family is and always has been the backbone of my strength, and I am very grateful to 
have them in my life. Lastly, I am greatly indebted to Darrin Masters for his patience, 
assistance, love, and support. It made all the difference. 



vi 



TABLE OF CONTENTS 

page 

ACKNOWLEDGMENTS iv 

LIST OF TABLES x 

LIST OF FIGURES xi 

ABSTRACT xii 

CHAPTER: 

1. INTRODUCTION 1 

Nutritional Ecology 2 

Calcium 3 

Previous Work in Bat Nutrition 4 

Bats and Flight 4 

Reproductive Costs 5 

Relieving Mineral Deficiencies 7 

Dissertation Focus 7 

2. RESEARCH STUDY SITES AND SPECIES 10 

Flying Foxes 10 

American Samoa 12 

Pteropus tonganus 15 

Papua New Guinea and Kau Wildlife Area 17 

Blossom bats of Papua New Guinea 18 

Syconycteris australis 19 

Macroglossus minimus 20 

The Lubee Foundation, Inc 21 

Pteropus vampyrus 21 

3. FRUIT CHOICE AND CALCIUM BLOCK USE BY TONGAN FRUIT BATS: DO 
FRUIT BATS SEEK OUT CALCIUM IN THEIR DIET? 24 

Introduction 24 

Methods 25 

Results 27 

Choice of High-Calcium or Low-Calcium Fruits 27 



vn 



Use of the Calcium Blocks 28 

Sugar 29 

Calcium 29 

Discussion 31 

4. FOLIVORY IN FRUIT BATS: ARE LEAVES A NATURAL CALCIUM 
SUPPLEMENT? 34 

Introduction 34 

Methods 36 

Results 39 

Discussion 41 

5. BIOAVAILABILITY AND APPARENT ABSORPTION OF MINERALS 
CONSUMED BY WILD TONGAN FLYING FOXES IN AMERICAN SAMOA 46 

Introduction 46 

Methods 49 

Netting and Housing of Bats 49 

Mineral Metabolism Experiments 50 

Analysis of Samples 52 

Statistical Analysis 53 

Results 53 

Mineral Consumption 53 

Mineral Absorption 56 

Discussion 57 

Bioavailability and Absorption 59 

Mineral Stress 61 

Future research 63 

6. NUTRITIONAL LANDSCAPE ECOLOGY AND HABITAT USE BY TONGAN 
FLYING FOXES IN AMERICAN SAMOA 66 

Introduction 66 

Methods 68 

Major Vegetation Types 68 

Nutritional Classification of the Major Vegetation Types 69 

Netting of Bats 70 

Radiotelemetry 70 

Radiotelemetry Error 72 

Results 73 

Radiotelemetry Error 73 

Habitat Selection 73 

Distance Traveled from Roost to Foraging Site 76 

Discussion 80 

Habitat Preference 80 

Distance Flown from Roost to Foraging Site 81 



vin 



Foraging Distance and Roost Affiliation 82 

Foraging Patterns of Reproductive Female Bats 83 

Nutritional Landscape Ecology 83 

7. ABSORPTION AND UTILIZATION OF MINERALS CONSUMED BY CAPTIVE 
LACTATING FEMALE MALAYAN FLYING FOXES {Pteropus vampyrus) AND 
THEIR PUPS 86 

Introduction 86 

Methods 88 

Results 91 

Discussion 95 

8. SUGAR CONCENTRATION PREFERENCES OF TWO SPECIES OF BLOSSOM- 
BATS {Syconycteris australis AND Macroglossus minimus) IN PAPUA NEW GUINEA 100 

Introduction 100 

Methods 103 

Nectar Concentrations 103 

Preference Tests 104 

Results 106 

Discussion 108 

9. CONCLUSION AND CONSERVATION RECOMMENDATIONS 114 

Testing the Calcium-Constraint Hypothesis 1 14 

Mineral Compensation 115 

Tongan and Malayan Fruit Bat Dietary Choice 116 

Summary 117 

Conservation Recommendations 118 

Major Threats to Fruit Bats 118 

The National Park of American Samoa 118 

Reducing Hunting Pressure and Bat Education Programs 1 19 

LIST OF REFERENCES 121 

BIOGRAPHICAL SKETCH 140 



IX 



LIST OF TABLES 
Table page 

3-1. Conjoint analysis results for all bats 29 

3-2. Conjoint analysis results for all females 29 

3-3. Frequency of calcium block use by Tongan fruit bats 30 

4-1. Age and gender of bats used in the experiment 38 

5-1. Results of the principal components analysis 54 

5-2. A comparison of nutrient levels offered and consumed 55 

5-3. A comparison of mineral apparent absorption values 56 

5-4. Mineral apparent absorption values for selected monogastrics species 58 

5-5. Apparent absorption values for minerals in animals under nutritional stress 58 

6-1 Nutrient classification of habitat types in American Samoa 70 

6-2. Summary of goodness-of-fit tests for habitat selection 75 

6-3. Data summary for the 20 Tongan radio collared fruit bats in American Samoa 77 

7-1. Composition of food fed to P. vampyrus at the Lubee Foundation, Inc 89 

7-2. A comparison of lactating female P.vampyrus with pups to nonbreeding 92 

7-3. A comparison of the diet offered and consumed by P. vampyrus 93 

7-4. A comparison of the mineral amounts for diet, orts, ejecta, mineral intake, 95 

8-1. Bats used in the nectar preference tests 106 



x 



LIST OF FIGURES 

Figure page 

3-1. Calcium block use by male, female and reproductive female Tongan fruit bats 30 

4-1. Supplemental calcium ingested (mg/g) by folivory for adult and juvenile males and 
females 39 

4-2. A comparison of total calcium ingested among habitual, occasional, and non-leaf- 
eaters 41 

4-3. Typical pattern of leaf consumption by Pteropus tonganus in American Samoa 41 

6-1. Map of Tutuila, American Samoa showing the three island habitat types 71 

6-2. Frequency of P. tonganus location estimates 74 

6-3. Percent of locations by habitat type on Tutuila, American Samoa 75 

6-4. A map of Tutuila, American Samoa showing the mean distance flown 79 

8-1. Distribution of blossom bats in Meganesia 101 

8-2. Banana flower nectar concentrations 104 

8-3. Results of 15% or 30% nectar-preference test 107 

8-4. Results of 15% or 7% nectar-preference test 107 

8-5. Average amount of nectar consumed over four nights for each age class 108 



XI 



Abstract of Dissertation Presented to the Graduate School 

of the University of Florida in Partial Fulfillment of the 

Requirements for the Degree of Doctor of Philosophy 

NUTRITIONAL ECOLOGY OF OLD-WORLD FRUIT BATS: 
A TEST OF THE CALCIUM-CONSTRAINT HYPOTHESIS 

By 

Suzanne Linn Nelson 

May 2003 

Chair: Dr. Stephen R. Humphrey 

Co-Chair: Dr. Tom Kunz 

Department: Wildlife Ecology and Conservation 

Although most studies argue that energy or protein is the most limiting 
component of the fruit bat diet, the calcium-constraint hypothesis proposes that 
reproduction in bats may be constrained by calcium rather than energy. 

To test this hypothesis, I experimentally tested if bats attempted to increase their 
calcium ingestion through preferential selection and consumption of calcium-rich foods. 
This research used four different species of fruit bats in three different geographic 
locations to test this theory. The various methods used included mineral metabolism 
trials, fruit choice and mineral block experiments, nectar concentration trials, and studies 
of habitat use by radiotelemetry. 

Results indicated that fruit bats seemed to base their food choices on the sugar 
content of fruits rather than the calcium content. Fruit bats preferred high-sugar 
agricultural fruits in all experiments, but bats did not meet their mineral requirements by 

xii 



consuming them. To compensate for mineral deficiencies resulting from reproduction or 
rapid growth, fruit bats may demonstrate an additional preference for concentrated 
mineral sources. Reproductive females and subadult bats appeared to select for additional 
calcium by consuming leaves and by using the calcium blocks. Future work should 
examine larger numbers of reproductive females and should observe bat foraging 
throughout the entire night to look for temporal deviations in resource use. Population 
persistence of Tongan fruit bats through time will reveal if these dietary choices are 
adaptive or maladaptive. 

Fruit bats act as seed dispersers and pollinators and are considered keystone 
species on isolated oceanic islands. Hunting, hurricanes, and habitat loss threaten bat 
populations. Native forest roosts are an essential resource component of the landscape for 
bat populations. This study suggests envisioning a nutrient landscape and then evaluating 
its components as they contribute to the needs of the bat population. A full explication 
will provide for conservation planning. 



xin 



CHAPTER 1 
INTRODUCTION 

Old- World fruit bats, or flying foxes, opportunistically feed on a wide variety of 
plants. Their diet includes fruits, flowers, leaves, shoots, buds, nectar, and pollen of 
tropical forest trees and shrubs (Start and Marshall 1974, Marshall 1985, Pierson and 
Rainey 1992, Wiles and Fujita 1992, Kunz and Diaz 1995, Banack 1996, Bonaccorso 
1998, Tan et al. 1998). Current estimates are that flying foxes consume the fruit of 136 
genera, flowers of 97 genera, and leaves of 10 genera (Courts 1998). Food choice may be 
influenced by a myriad of factors, including energy needs, requirements for specific 
nutrients, reproductive status, constraints of the digestive system, abundance, diversity, 
and seasonality of different food items, and competition and predation (Fleming 1988, 
Oftedal 1991). 

Fruit bats in the wild appear to meet their nutrient needs by consuming large 
quantities of a wide variety of native fruits (Dempsey 1999). Their food tends to be 
conspicuous, abundant, and easily harvested within clumps (Mickleburgh et al. 1992). 
Hundreds of bats may descend on a locally and temporarily abundant food source until it 
is depleted (Pierson and Rainey 1992, Wilson and Engbring 1992). Fruit bats are often 
regarded as "sequential specialists," favoring preferred resources among a group of foods 
as they become seasonally available (Marshall 1983, 1985, Banack 1998). 

The largest genus of the Old World fruit bats, Pteropus, is primarily an island- 
dwelling taxon, with 97% having some or all of their distribution on islands 
(Mickleburgh et al. 1992, Pierson and Rainey 1992). Fruit bats are crucial to establishing 

1 



and maintaining forest composition on isolated oceanic islands that have a 
characteristically limited suite of animal pollinators and dispersers (Thornton et al. 1990, 
Cox et al. 1991, Elmqvist et al. 1992, Cox and Elmqvist 2000, Cook et al. 2001). Flying 
foxes are highly mobile and can travel 40-60 km to reach a feeding area (Marshall 1983, 
Rainey et al. 1995, Banack 1996). Thus, they are able to transport seeds great distances as 
they drop or defecate them while in flight (Rainey et al. 1995, Shilton et al. 1999). Fruit 
bats are particularly important to tropical forest regeneration following natural 
catastrophes or forest destruction by humans (Bonaccorso and Humphrey 1978, 
Whittacker and Jones 1994, Thornton et al. 1996,), and they can influence the 
composition and distribution of food resources within the landscape (Kunz 1996). 
Although contributions of fruit bats to tropical forests are well documented, the factors 
that influence food choice are still highly debated and largely unknown. Factors that 
influence food choice can be evaluated within the discipline of nutritional ecology. 

Nutritional Ecology 
Nutritional ecology includes the study of organic and mineral nutrients. Organic 
nutrients consist of fiber, carbohydrates (fiber and soluble carbohydrates), protein 
(nitrogen), fat and/or energy. Mineral nutrients are inorganic elements and include 
calcium, phosphorus, and iron. Minerals required in gram quantities in the body are 
referred to as macrominerals and consist of calcium, phosphorus, sodium, chlorine, 
potassium, magnesium, and sulfur. Macrominerals are important structural components 
of bone and other tissues and play vital roles in the maintenance of acid-base balance, 
osmotic pressure, and membrane electrical potential (McDowell 1992, NRC 2001). 
Minerals required in milligram or microgram amounts are referred to as trace minerals. 
This group includes copper, iodine, iron, manganese, molybdenum, selenium, zinc, and 



perhaps chromium and fluorine. Trace minerals are present in very low concentrations 
and often serve as components of metalloenzymes, enzyme cofactors, or hormones in the 
endocrine system (NRC 2001). Studies of dietary minerals quantify and compare mineral 
concentrations in foods consumed by animals. The study of minerals, particularly 
calcium, represents the primary focus of my research. 

Calcium 
About 98% of the calcium in the body is located within the skeleton (McDowell 
1992). Calcium is essential for the formation of skeletal tissues, transmission of nervous 
tissue impulses, excitation of skeletal and cardiac muscle contraction, blood clotting, and 
as an important component of milk (NRC 2001). The calcium concentration of plasma 
must be maintained at a relatively constant value of 1 to 1 .25 mM to ensure normal nerve 
membrane and muscle end plate electric potential and conductivity (NRC 2001). 
Vertebrates have evolved an elaborate system to maintain calcium homeostasis whenever 
there is a loss of calcium (NRC 2001). The parathyroid glands monitor the concentration 
of calcium in carotid arterial blood. When levels drop, calcium can be replaced by 
resorption of calcium stored in bone, by increased calcium absorption, or by reducing 
urinary calcium loss (NRC 2001). Active transport of calcium, an energy-requiring 
process, appears to be the major route of calcium absorption and is controlled by the 
hormone form of vitamin D (Wasserman 1981, Bronner 1987, Holick 2002a, 2002b). If 
dietary calcium is severely deficient for a prolonged period, an animal can develop severe 
osteoporosis to the point of developing fractures (Radostits et al. 1994). Despite the 
skeletal changes that result from prolonged calcium deficiency, plasma calcium levels are 
homeostatically maintained and will only be slightly lower than normal (NRC 2001). 






Previous Work in Bat Nutrition 

Previous nutritional studies of fruit bats identified the organic nutrient protein 
(nitrogen) as the major limiting nutrient in the diet (Thomas 1984, Herbst 1986, Steller 
1986). It was previously thought that fruit bats consumed only fruits and did not 
supplement their diet with other foods (Thomas 1984, Herbst 1986). Yet fruits were 
considered nutrient-poor because of their low fat and protein content (Mattson 1980, 
Herrera 1987, Witmer 1998). Daily fruit intake and preference were thought to be 
dictated by protein rather than energy content of the diet. It was thought that bats over- 
consumed energy to obtain adequate protein (Thomas 1984) 

Later research showed that the diet of fruit bats includes much more than just fruit 
(Banack 1996, Courts 1998). Fruit bats have been reported as deliberately ingesting 
insects, pollen, and leaves, possibly to provide extra protein in the diet (Drew 1988, 
Mickleburgh et al. 1992, Kunz and Diaz 1994, Kunz and Ingalls 1995, Kunz 1996, 
Courts 1998). Thus, protein may not be a limiting nutrient in the bat diet. Instead, 
minerals such as calcium may be deficient in the diet of bats (Barclay 1995). Barclay 
(1994, 1995) proposed that because of bat's adaptation to flight, reproduction of females 
bats is most constrained by their intake of calcium rather than their intake of protein or 
energy. 

Bats and Flight 

The unique adaptation of flight in bats imposes certain restrictions on their 
reproduction and development not experienced by other mammals. The wing skeleton of 
a growing bat must acquire structural and material characteristics that will enable it to 
withstand the mechanical pressures of flapping flight (Bernard and Davison 1996). This 
includes adequate mineralization to confer strength and stiffness and to resist torsional 



and/or bending stresses (Papadimitriou et al. 1996). Bats and similarly sized terrestrial 
mammals produce litters with a mass averaging 25% that of the female (Kurta and Kunz 
1987, Hayssen and Kunz 1996). Compared to other mammals, bats raise their young to a 
significantly larger size because young cannot fly and gain independence until they are 
almost fully grown (Barclay 1995). Juvenile bats are unable to fly or forage 
independently until they have achieved approximately 70% of adult mass and more than 
95% of adult skeletal size (Kurta and Kunz 1987, Barclay 1995,). 

Rodents typically produce small litters that are weaned quickly at 30-44% of adult 
size and obtain some of their nutrition by foraging for themselves (Millar 1977, Barclay 
1995). In contrast, maternal milk is the only energy and nutrient source for dependent 
young bats. Juvenile flying foxes associate with their mothers for up to a year (Kurta and 
Kunz 1987, Mickleburgh et al. 1992, Pierson and Rainey 1992) and will opportunistically 
nurse if in close proximity, especially in captivity (D. LeBlanc, pers. comm.). Thus, near 
the end of lactation, females provide total nutrition to bat offspring that are nearly of 
adult size (Kunz et al.1995, Kunz and Stern 1995, Hood et al. 2001). Overall, bat pups 
are more expensive, in terms of energy and nutrients, to a female bat than is each young 
to an equivalently sized terrestrial mammal (Barclay 1995). Each young requires a large 
parental investment which may restrict the total number of young that can be raised 
(Barclay 1994, Kunz and Hood 2000). 

Reproductive Costs 

Several studies indirectly addressed the question of calcium demand during 
pregnancy and lactation, and showed that these are periods of calcium stress for bats 
(Kwiecinski et al. 1987a, 1987b, Studier et al.1991, Sevick and Studier 1992, Studier et 
al. 1994a, 1994b, Bernard and Davison 1996). Nutritional requirements for females 



increase dramatically during reproduction, and females may be in negative calcium 
balance from the onset of pregnancy to the end of lactation (Bernard and Davison 1996). 

Females bats bear almost the entire mineral cost of raising their offspring by 
allocating their own skeletal calcium reserves to build the skeletons of their young 
(Bernard and Davison 1996, Papadimitriou et al. 1996). The bones of females become 
more porous as stored calcium is depleted during prolonged lactation (Sevick and Studier 
1992, Radostits et al. 1994). The excessive calcium demands of raising several young in 
sequential years can result in osteoporosis in female bats, particularly in the mandible and 
the long bones of the wings (Kwiecinski et al. 1987a). Through time, this has the 
potential to decrease a female's fitness. The increased risk of wing-bone fractures can 
impede her ability to fly and forage. Tooth loss and the subsequent inability to chew 
fruits and leaves could affect longevity, fitness, and overall health (Barclay 1995). 

Keeler and Studier (1992) found that among reproductive female bats, all caloric 
requirements were met, but calcium intake was one-tenth the estimated requirement. For 
a lactating female, inadequate calcium can result in low milk production (McDowell 
1992). Several factors can influence the postnatal growth of mammals, including age, 
nutritional and hormonal condition of the mother, and milk quality and quantity (Hoying 
and Kunz 1998, Kunz and Hood 2000). For nursing bat pups, inadequate milk results in 
inhibited growth and reduced mineralization of bone, which can result in lameness and 
bone fractures (Radostits et al. 1994, NRC 2001). Studies of other mammals showed that 
inadequate quantities of calcium in the diet can affect fecundity, number of litters, and 
survival of offspring (Batzli 1986, Delgiudice et al.1990). Calcium maybe a limiting 
nutrient for reproductive bats that could influence population density through time. 



Relieving Mineral Deficiencies 

Bats may be able to delay or reduce the effects of calcium deficiency by feeding 
on calcium-rich foods or on concentrated calcium sources. For example, O'Brien et al. 
(1998) proposed that frugivores eat figs because they have higher calcium concentrations 
than many other native and agricultural fruits. Calcium concentration of figs in American 
Samoa were three times that of other native fruits, and over ten times more concentrated 
in calcium than agricultural fruits on the island (Nelson et al. 2000a). Leaves are also 
especially rich in calcium and are consumed by bats (Lowry 1989, Kunz and Ingalls 
1994, Kunz and Diaz 1995,Tan et al 1998, Ruby et al. 2000). Concentrated sources of 
calcium can supplement the diet of fruit bats, and maybe important in times of greater 
physiological need such as pregnancy and lactation. 

Dissertation Focus 

Studies of female food choices that result from the energetic and mineral demands 
of pregnancy and lactation are limited. Robert Barclay proposed bold new ideas on what 
may motivate female food choice during reproduction (Barclay 1994, 1995). My research 
tests Barclay's "calcium-constraint hypothesis" which proposes that reproductive females 
are more constrained by calcium than energy in their diet (Barclay 1994, 1995). I 
examined whether bats attempted to increase their consumption of deficient nutrients 
through dietary choice. 

It is a challenge to conduct nutrition work on Old World fruit bats because the 
requirements for calcium and all other minerals needed to maintain bat health are 
currently unknown. Previous dietary work most often used mammalian standards or 
standards for rats to determine bat requirements, despite different diets, digestive systems 
and metabolic rates of bats (Oftedal and Allen 1996). Thus, one of the goals of my 



8 

research was to establish mineral requirements for flying foxes that could be used for 
future wild and captive nutrition studies. By offering free-choice diets to fruit bats and 
quantifying mineral consumption, I hoped to create a mineral profile of bat feeding. 

This research focused on three different geographic regions and four different fruit 
bat species that represent three different size classes of bats. Each species was appropriate 
to answer a specific question on fruit bat feeding and nutrition. Research was performed in 
Papua New Guinea on the tiny fruit bats (18-20 g) Macroglossus minimus and 
Syconycteris australis, in American Samoa on the mid-sized fruit bat (300-600 g), 
Pteropus tonganus, and on one of the world's largest fruit bats (1000-1500 g), Pteropus 
vampyrus, at the Lubee Foundation, Inc., in Gainesville, Florida. 

Most of the nutritional work in my research involved the Tongan fruit bat in 
American Samoa. This species was chosen because it is mid-sized (300-600 g), is one of 
the most widely distributed of all Pteropus species (Koopman and Steadman 1995, Miller 
and Wilson 1997), and it feeds in both native and agricultural habitats (Pierson and 
Rainey 1992, Banack 1996). Together, these features suggest that this species is highly 
adaptable and is an excellent general model. Therefore, results of nutritional studies in 
this species should be widely applied to many other flying fox species. The Tongan 
flying fox was used to generate new information on wild fruit bat mineral levels, leaf 
consumption, use of concentrated mineral sources, and habitat use while foraging. 

One of the world's largest flying foxes, Pteropus vampyrus, was used to study 
mineral consumption in a population of captive lactating females and their dependent 
pups. The large size of P. vampyrus may result in excessive mineral requirements as 
females attempt to meet the nutrient demands of a large pup for up to a year. Thus, 



quantifying the mineral content of the diet that can meet the requirements of this large 
flying fox can be helpful in formulating diets for reproductive females of other captive 
flying fox species. Mineral absorption values can also be compared to determine if wild 
Tongan fruit bats are mineral deficient. 

Because Barclay (1995) argued that reproductive females were more constrained 
by calcium than energy, I examined energy consumption in two species of blossom bats in 
Papua New Guinea. These species are among the smallest members of the Old World fruit 
bats. They have a high metabolic rate and little or no means of storing large energy 
reserves (Lemke 1984), so they may be energy-sensitive foragers limited by food 
availability (Law 1992, 1993b, 1994a). Different nectar solutions representing different 
energy concentrations were given to the bats in choice trials to determine whether energy 
was a limiting nutrient for these species. 

I used two methods new to the study of bats. To quantify bat mineral 
consumption, I used apparent absorption, a method commonly used in animal science but 
rarely in wildlife ecology because of the need to use captive animals. This method 
accounts for minerals in the both the food and feces, and for minerals lost in the fibrous 
pellet expelled while fruit bats feed (Lowry 1989, Kunz and Ingalls 1994). Also new to 
the study of bats is the idea of nutritional landscape ecology, which examines whether 
bats use mineral-rich habitats while foraging throughout the landscape. 

For all these studies, I predicted that bats will seek out calcium-rich sources to 
supplement their diet. The additive effect of the constraints of flight, reproductive costs, 
and calcium-poor diets should make calcium consumption a priority. 



CHAPTER 2 
RESEARCH STUDY SITES AND SPECIES 

Flying Foxes 

The Order Chiroptera consists of two suborders: Microchiroptera and 
Megachiroptera. Current estimates suggest that Megachiroptera separated from 
Microchiroptera 50.2 million years ago (Bastian et al. 2001). Unlike Microchiroptera, 
which has a global distribution, members of Megachiroptera, also called Old World fruit 
bats, are limited to the Paleotropics. Megachiroptera comprise the single family 
Pteropodidae, which contains 42 genera and 191 species (N. Simmons, pers.comm.). 
Flying foxes are members of the genus Pteropus, which contains 58 species (Koopman 
1993). Almost 97% of flying fox species are island dwelling, and 35 species are confined 
to a single island or island group (Mickleburgh et al. 1992). Members of the genus 
Pteropus range from Madagascar to India, and from Southeast Asia to Australia, and 
reach as far east as the Cook Islands (Pierson and Rainey 1992). 

The genus Pteropus feeds on fruit, leaves, nectar and pollen of trees found within 
both native forests and agricultural areas. Across their range, Pteropus species are known 
to visit over 92 genera of plants in 50 different plant families (Marshall 1985, Wiles and 
Fujita 1992). Fruit bats are important pollinators and seed dispersers in tropical forest 
ecosystems (Fleming 1988, Rainey et al. 1995, Banack 1996). On many isolated oceanic 
islands with depauperate pollinator and seed disperser faunas, flying foxes are the only 
animals capable of carrying large-seeded fruits. In island ecosystems of the south Pacific, 
flying foxes are considered keystone species because their extinction could result in a 

10 



11 

significant decline in both native forest diversity and regeneration (Cox et al. 1992, 
Rainey et al. 1995). Their role as long distance seed dispersers (Shilton et al. 1999) 
further demonstrates their critical role in maintaining forest structure and integrity. Seed 
dispersal by bats may increase seed survival by decreasing seed predation and increasing 
the chances of landing in favorable microhabitats (Jantzen et al. 1976, Augspurger 1983). 

Flying foxes are phytophagus and consume fruits, pollen, flowers, nectar and 
leaves of plants (Marshall 1985). They process fruit by pressing the tongue against the 
palate to break up the fruit and ingest the pulp and juices. Flying foxes also consume 
leaves as a regular part of their diet (Kunz and Ingalls 1994, Kunz and Diaz 1995). They 
chew the leaves and swallow the juice, ejecting the fiber portion as a small compressed 
pellet (Lowry 1989). The ingestion of mostly the liquid portion of fruits and leaves 
results in a food transit time as low as 20 minutes (Tedman and Hall 1985). The 
advantage of this foraging strategy to a flying mammal is reduced bulk and wing loading, 
and reduced energy expenditure traveling to foraging areas (Kunz and Ingalls 1994). 

Most flying fox species give birth to one offspring per year starting when they are 
two years old (Pierson and Rainey 1992). Gestation periods range from 4-6 months 
followed by a rearing interval of equal length, although young may stay associated with 
their mothers for up to a year (Banack 1996, Hall and Richards 2000). A few species (P. 
mariannus yapensis in Yap, P. molossinus in Pohnpei, and P. tonganus in Samoa) may 
have more than one birth peak per year (Pierson and Rainey 1992). Birth peaks are often 
correlated with seasonality and resource availability (Heideman 1995, Racey and 
Entwistle 2000). Bats of the genus Pteropus typically give birth to a single young, but 
twinning can occur. 



12 

Habitat alteration and habitat loss are the primary reasons for declining 
populations of flying foxes (Cheke and Dahl 1981, Fujita and Tuttle 1991, Mickleburgh 
et al. 1992, Pierson and Rainey 1992). Many species, particularly those inhabiting 
mangrove swamps and lowland forest have lost critical roosting areas (Mickleburgh et al. 
1992). Fruit bat populations are also heavily influenced by human depredation and 
tropical storms (Wodzicki and Felton 1980, Cheke and Dahl 1981, Heaney and Heideman 
1987, Craig et al. 1994a, 1994b, Loebel and Sanewski 1987). 

American Samoa 

The Samoan islands are a biogeographical unit politically divided into American 
Samoa, an unincorporated territory of the United States, and Samoa, a sovereign country. 
They lie in the South Pacific Ocean (14° S, 170° W), approximately 4000 km southwest 
of Hawaii and 3000 km northwest of New Zealand. American Samoa comprises Tutuila, 
Aunu'u, the Manu'a Islands (Ta'u, Olosega, and Ofu), Rose Atoll, and Swains Island. 
The largest island in American Samoa is Tutuila, which is 142 km 2 in area, supports 90% 
of the human population, and contains the capital village of Pago Pago (Craig and Syron 
1992). 

The islands are volcanic in origin, having risen from hot spots on the ocean floor 
in the late Pliocene or early Pleistocene (Kear and Wood 1959). The islands are now 
highly eroded, resulting in extremely steep topography and deeply cut valleys. The 
climate is warm and humid year-round and considered moist tropical, with an average 
annual temperature of 25°C. There are two distinct seasons, the wet and dry season, 
although rain falls 300 days/year (Amerson et al. 1982). The wettest months are October 
through March and rainfall averages 3200 mm annually (NOAA 1996). 



13 

Samoa lies within the South Pacific hurricane belt and is subject to hurricanes and 
tropical storms. The Samoan archipelago was battered by three intense storms in 1986, 
1990, and 1991 . These were the most severe storms to occur in Samoa in over 160 years, 
with each sustaining winds between 200 km/h and 240 km/h (Elmqvist et al. 1994). 
Because the storms occurred so close together in time, they resulted in extensive damage 
to native forest trees. Fruit bat populations were severely decimated due to loss of food, 
roosts, and increased hunting immediately after the storms (Craig et al. 1994b, Elmqvist 
et al. 1994, Pierson et al. 1996, Hjerpe et al. 2001). A sharp rise in air temperature over 
the past decade (NOAA 1999) suggests further climatic uncertainty and a probable 
increase in the frequency of hurricanes in the area (Craig et al. 2000). 

Pacific island archipelagos display very high levels of endemism; typically 30%- 
50% of the plants occur nowhere else (Brautigam and Elmqvist 1990, Cox et al. 1992). 
More than 326 genera of vascular plants can be found in the Samoan archipelago 
(Christophersen 1935) and at least 68 are endemic plant species (Amerson et al. 1982). 
The affinities of most plants are Australian or Malesian (Whistler 1992). Paleotropical 
rainforest is the natural vegetation of Samoa. 

Rainforest originally covered nearly the entire surface of Samoa. Mature rain 
forest is now restricted to the least accessible areas such as steep interior slopes and the 
wet, cool montane regions away from villages. Rainforest is typically a tall forest with a 
canopy up to 30 m in height (Whistler 1992). Rainforest trees in American Samoa 
produce fleshy fruits adapted for dispersal by frugivorous birds and bats (Freifeld 1998). 
The four types of rainforest are coastal, lowland, montane, and cloud forest, and each is 
most easily distinguished by the plant species found within them (Whistler 1992). 









14 

Anthropogenic disturbances such as shifting cultivation and the development of 
agroforestry have replaced much rainforest with secondary forest and agricultural lands. 
Secondary forest is less diverse than mature rainforest, and is dominated by shade- 
intolerant trees that quickly establish in disturbed areas (Whistler 1992 Freifeld 1998). 
Cultivated lands consist of local slash-and-burn plots growing together in a mix of forest 
and agricultural plants and are found in valleys and near villages (Cole et al. 1988). 

The flora of American Samoa includes over 800 species of angiosperms that are 
pollinated by 1 1 species of wasps, 9 species of frugivorous/nectarivorous birds, and 2 
species of phytophagus bats (Cox et al. 1992). Other animals on the islands include 
snakes, skinks, toads, and large land snails. Rats, pigs, cows, dogs, and cats have been 
introduced by humans. Feral pigs threaten native forest trees by destroying tree bark and 
disturbing seed banks of native species. Feral cats penetrate secondary and agro forest 
areas and prey on resident and migratory birds (S. Nelson, pers.obs.). 

The Samoan islands have been inhabited by humans for approximately 3,000 
years (Kirch and Hunt 1993, Petchey 2001). Today, the most serious environmental and 
social problem facing American Samoa is its rapid human population growth (Craig et al. 
2000). The population estimate of 63,000 for the year 2000 is increasing at a rate of 
approximately 2.5% annually, which will result in a doubling time of only 28 years. A 
continued increase is expected given the high birth rate (4.5 children per female) and high 
proportion of pre-reproductives in the population; nearly 50% of the population is less 
than 20 years old (Craig et al. 2000). This exponential increase in the human population 
is coupled with a substantial degree of habitat loss due to the confines of a small oceanic 
island. Annually, 1-2% of rainforest is lost to agroforest in Samoa (Cole et al. 1988), 



15 

although current rates of forest loss may be higher. Land-use practices on Tutuila are 
largely influenced by the steep topography of the island. With 50% of the land area 
having a slope greater than 70%, there is relatively little land to use for agriculture and 
housing (Craig et al. 2000). As the human population expands, hillsides are being 
developed as are areas that were not formerly considered. 

One area of refuge from development and human encroachment is the National 
Park of American Samoa. The National Park of American Samoa was officially 
established in September 1993 when a 50-year lease was signed between the National 
Park Service and the American Samoa Government representing the villages in the Park. 
The 9,000 acre park is spread out over three islands (Tutuila, Ta'u, Ofu), and includes 
1000 acres that are underwater. The National Park of American Samoa contains many 
critical roosting sites for both species of fruit bats (Brooke 1998, Brooke 2001) and 
provides valuable native habitat for other species on the island (Freifeld 1999). 

Pteropus tonganus 

Three species of bats occur in the Samoan archipelago, Pteropus samoensis, 
Pteropus tonganus, and Emballanura semicaudata (Peale 1848, Andersen 1912). E. 
semicaudata is currently on the verge of local extinction in Samoa (Grant et al. 1994) or 
may already be extinct. The two pteropodid bat species found on the Samoan islands are 
the Samoan flying fox {Pteropus samoensis Peale), and the Tongan flying fox {Pteropus 
tonganus Quoy and Gaimard). The Samoan flying fox is solitary, diurnal, and prefers to 
forage on native forest fruits (Wilson and Engbring 1992, Thomson et al. 1998, Brooke et 
al. 2000, Brooke 2001). In contrast, the Tongan flying fox roosts in noisy colonies of 
thousands of bats, is primarily nocturnal, and forages both in native forest and 
agricultural areas (Quoy and Gaimard 1830, Banack 1996). Also called the white- 



16 

collared or insular flying fox, P. tonganus is a medium-sized fruit bat that weighs 
between ca. 300-600g, with a forearm length of 120-160 mm (Miller and Wilson 1997). 
Males are generally larger than females (Flannery 1995b). The fur is black or seal brown 
with a contrasting creamy yellow mantle (Miller and Wilson 1997). 

The Tongan flying fox is common throughout the South Pacific and has the 
largest geographic range of any pteropodid species in Oceania. It is one of the most 
widely distributed of all Pteropus species (Koopman and Steadman 1995) and is found 
south of the equator from the Schouten Islands off NE New Guinea, south to New 
Caledonia, and east to the Cook Islands (Koopman 1993). The distribution of P. tonganus 
includes the easternmost limit for the Pteropodidae. The Tongan flying fox is described 
by Koopman (1979) as a "supertramp" species, referring to its absence from the largest 
and most species-rich islands and its prevalence on small, species-poor ones (Pierson and 
Rainey 1992, Mickleburgh et. al. 1992). There are few morphological differences among 
populations of P. /. tonganus separated by several hundred kilometers, although animals 
from Niue and the Cook Islands tend to be slightly smaller (Wodzicki and Felton 1980). 

Tongan fruit bats roost colonially. Colonies can range in size from several 
individuals to several thousand. Roosts in coastal forest are in undisturbed areas on steep 
slopes immediately above the ocean or in upland forest (Brooke 1998). Within the roost, 
bats hang together in harem groups consisting of a single dominant adult male with two 
to sixteen adult females and their young, but group composition is highly labile (Banack 
1996). Tongan fruit bat births have been observed year round in Samoa. The high number 
of pregnant and lactating females throughout the year suggests the ability of this species 
to rapidly increase its population size under optimal conditions (Banack 1996). The 



17 



mother carries her young until they are able to fly at 2-3 months of age (Brooke 1999). 
About 50% of copulations observed involved females that were nursing young, 
suggesting postpartum estrus in this species (Banack 1996). 

Tongan fruit bats disperse from colonies to forage in native forests, agricultural 
areas, and residential areas (Banack 1996). They are described as favoring agricultural 
areas, but the extent of use for each habitat type is currently unknown (Brooke 1998). 
They are phytophagus and visit numerous plants to consume fruit, nectar, pollen and 
leaves. A highly plastic forager, Pteropus tonganus uses 42 species of plants in American 
Samoa and has the ability to find food despite seasonal and distributional changes in food 
availability (Elmqvist et al. 1992, Banack 1998). 

Mortality among Tongan flying foxes includes predation from raptors and snakes, 
disease epidemics, hurricanes, hunting by local people, and habitat loss (Brooke 1998). In 
many Pacific islands, fruit bats are considered a delicacy and consumed by local people, 
and were previously hunted to supply a luxury food trade (Craig et al. 1994a, Wiles et al. 
1997). Archeological records indicate that P. tonganus has been hunted and eaten for at 
least the last 1,000 years (Steadman and Kirch 1990). In American Samoa, a ten year 
hunting ban was enacted to limit depredation (Craig and Syron 1992, Craig et al. 1994b, 
Brooke 2001). A current population estimate is 6300 or more Tongan fruit bats in 
American Samoa (A.P. Brooke, pers. comm.). Pierson et al. (1992) listed this species as 
priority grade II (not threatened) with priority status unknown. 

Papua New Guinea and Kau Wildlife Area 
Papua New Guinea comprises over 600 islands and includes the eastern half of 
the island of New Guinea, the two northernmost islands of the Solomon chain 
(Bougainville and Buka), and the Bismarck and Admiralty archipelagos. Papua New 



18 

Guinea lies between the Coral Sea and the South Pacific Ocean and is north of Australia. 
The country is a member of the British Commonwealth with Port Moresby as its capital. 
It is home to approximately five million people that speak over 800 languages. 

Of the estimated 15,000 to 21,000 vascular plants found in Papua New Guinea, 
more than half are believed to be endemic (Mittermeier et al. 1997). Similarly, mammals 
are very diverse, with 242 species represented, 57 of which are endemic (Bonaccorso 
1998). Bats, rodents and marsupials account for the bulk of the mammalian diversity, but 
bat species are the most numerous and highly diverse. Thirty- four species of Old World 
fruit bats are found in Papua New Guinea. 

Fieldwork was done in the northern province of Madang in the Kau Wildlife 
Area. The forest at Kau Wildlife Area is an old and relatively undisturbed lowland 
rainforest tract of 300 ha that is an owned by the Dipida Clan. This area has been 
untouched by logging, shifting cultivation, burning, or hunting with firearms since 1963 
when it was put aside by the clan for conservation. Laboratory work was conducted at the 
former Christensen Research Institute near Madang. 

Blossom bats of Papua New Guinea 
Both Syconycteris australis and Macroglossus minimus belong to the subfamily 
Pteropodinae (formerly Macroglossinae) (Kirsch et al. 1995) which reaches its maximum 
diversity in New Guinea (Flannery 1995a). Both S. australis and M. minimus have large 
geographic distributions and exhibit energetic plasticity. They are able to live in a variety 
of environments, including small islands, disturbed successional forests, lowlands, and 
montane rain forests (Bonaccorso and McNab 1997). S. australis is often described as a 
feeding and habitat generalist in Papua New Guinea, while M. minimus is a nectar 
specialist (Bonaccorso 1998) This is opposite of Australia, where S. australis is a nectar 



19 

specialist (Law 1992). Morphologically, the two species of bats are almost identical in 
linear size and body mass (approx. 15-20 g). They both have an elongated rostrum and a 
slender, protrusible tongue with brushlike projections to gather nectar and pollen. Their 
broad, short wings permit hovering and maneuverability (Gould 1978, Nowak 1999). 
Syconycteris australis 

The northern blossom bat, S. australis, resides in a variety of habitats, including 
lowland rain forest, dry sclerophyll woodland, montane/hill forest, and swamp forest 
(McKean 1972, Bonaccorso 1998). Its geographic distribution includes New Guinea and 
the east coast of Australia into New South Wales (Richards 1983, Bonaccorso and 
McNab 1997). This species weighs approximately 18 g and has an average forearm 
length of 39 mm (Law 1992, Bonaccoso 1998, Nowak 1999). 

During the day, S. australis roosts singly in the foliage of trees and shows fidelity 
to day roost areas (Law 1996, Winkelmann et al. 2000). Northern blossom bats are 
excellent thermoregulators and are able to inhabit almost the entire range of elevations in 
New Guinea (McNab and Bonaccorso 1995). S. australis has a field metabolic rate that is 
double that predicted for an animal its size but a basal metabolic rate lower than predicted 
for its size (Geiser and Coburn 1999). An ability to undergo short periods of torpor may 
explain its extended distribution range and counteract unpredictable nectar availability 
and extended day roosting (Geiser et al. 1996, Coburn and Geiser 1998, McNab and 
Bonaccorso 2001). 

S. australis is distinguished from M. minimus by a more robust dentition, which is 
better suited for feeding on fruits, but it also consumes nectar, pollen, and occasional 
insects (Bonaccorso 1998, Flannery 1995a). S. australis displays geographic variation in 
its diet within Australia. Law (1992, 1994) found S. australis to be a nectar specialist in 



20 

its range in southern Australia, but is a frugivore and partial folivore in the more northern 
part of its range. S. australis actively discriminated between sugar concentrations in 
concentration preference tests (Law 1993). Flower morphology may influence S. 
australis foraging, and result in resource partitioning within habitat areas (Nicolay and 
Dumont 2000). The northern blossom bat is abundant and ubiquitous in distribution 
throughout a wide range of habitat types and is listed as lower risk: least concern in the 
1996 IUCN Red List of Threatened Animals (Baillie and Groombridge 1996). 
Macroglossus minimus 

Macroglossus minimus, also called the southern blossom bat, is the smallest of the 
blossom bats found in New Guinea. This bat weighs only 12-18g and has a forearm 
averaging 37.5 mm (Flannery 1995a, Nowak 1999). It is similar morphologically to S. 
australis, but is unique in its feeding habits and more limited in its distribution. M. 
minimus is a nectar specialist that occurs from sea level to 1000 m (Flannery 1995a, 
Bonaccorso 1998). Specimens from New Guinea lacked fruit in the stomach (McKean 
1983), and captive individuals refused to eat fruit (Mickleburgh et al. 1992). Instead, M. 
minimus seems to prefer the nectar and pollen (Nowak 1999). 

M. minimus is widespread in lowland New Guinea, where it is often found in 
secondary rain forest, hill forests, or coastal areas near mangroves (Mickleburgh et al. 
1992, Bonaccorso 1998). It feeds in disturbed habitats and orchards due to its preference 
for domesticated bananas (Gould 1978, Bonaccorso 1998). This species roosts singly, in 
mother-infant pairs, or in small groups on the underside of large leaves, tree branches, or 
roofs of abandoned buildings (Bonaccorso 1998). Births can occur in any month, perhaps 
due to sperm storage (Hood and Smith 1989). 



21 

The southern blossom bat has a metabolic rate that is much lower than expected 
for a mammal of its body mass (Bonaccorso and McNab 1997). This species may be 
restricted to tropical areas because it can enter shallow daily torpor (Battels et al. 1998). 
M. minimus is abundant and widespread throughout its range and is not currently 
threatened (Mickleburgh et al 1992); the small size and cryptic roosting may contribute to 
its listing as low risk: least concern in the 1996 IUCN Red List (Baillie and Groombridge 
1996). 

The Lubee Foundation, Inc. 

The Lubee Foundation, Inc. was founded by the late Luis F. Bacardi in 1990 as a 
non-profit organization involved in the conservation of threatened and endangered 
species of Old World fruit bats. It maintains captive breeding populations and supports 
research of both captive and wild populations of bats at its facility near Gainesville, 
Florida. Bats are housed in outdoor circular flight exclosures that surround temperature- 
controlled roosting quarters. Eleven species of bats are currently housed at this facility, 
representing both fruit and nectar feeding bats. These species include Cynopterus 
brachyotis, Eidolon helvum, Epomophorous wahlbergi, Glossophaga soricina, Pteropus 
giganteus, Pteropus hypomelanus, Pteropus poliocephalus, Pteropus pumilus, Pteropus 
rodricensis, Pteropus vampyrus, and Rousettus aegypticus. In total there are over 600 
fruit bats currently housed at the facility. My research at The Lubee Foundation, Inc. 
focused on P. vampyrus. 

Pteropus vampyrus 

P. vampyrus, also called the Malayan flying fox, is one of the world's largest 
flying foxes, and can attain a wingspan of 2 m. Members of this species weigh between 
645-1092 g and have a forearm of 180-220 mm. P. vampyrus has a distinctively dog or 



22 

foxlike face, and pelage color ranges from mahogany or orange to black with a black, 
brown or grey/silver underbelly (Kunz and Jones 2000). The Malayan flying fox inhabits 
Burma, Thailand, the Phillipines, Sumatra, Java, Borneo, the Lesser Sundas, and adjacent 
islands (Andersen 1912, Medway 1969, Corbet and Hill 1992). There are currently seven 
recognized subspecies of P. vampyrus (Bastian et al. 2001). 

The Malayan flying fox is found in a variety of habitats including primary forest, 
mangrove forests, mixed fruit orchards, and coconut groves (Medway 1969, Payne et al. 
1985, Heideman and Heaney 1989). In Malaysia, P.vampyrus is most often found 
roosting in isolated and inaccessible areas such as mangrove forests and freshwater 
swamps (Payne et al.1985, Mohd-Azlan et al. 2001). Roosts are in the canopies of 
emergent trees and are often shared with A.jubatus. Mixed groups of these two species 
may range from 500 to 150,000 individuals (Mudar and Allen 1986, Heideman and 
Heaney 1989). Roosts abandonment is most often due to disturbance, habitat loss, or 
hunting (Mohd-Azlan et al. 2001). 

Malayan flying foxes fly up to 50 km to reach their feeding grounds and shift 
feeding sites in response to changes in food availability (Medway 1969). They feed on 
flowers, nectar, and fruit, but most often on flowers and nectar (Gould 1977, Goodwin 
1979, Payne et al. 1985). Pollen, nectar, and flowers of coconut and durian trees (Durio 
zizebethinus), fruits of rambutan (Nephelium lappaceum), figs (Ficus spp.) and langsat 
(Lansium domesticum) trees, in addition to fruits such as mangos and bananas are all 
preferred foods (Heideman and Heaney 1989, Kunz and Jones 2000) and are actively 
defended (Gould 1977). Figs are a dietary staple of P. vampyrus, while other foods are 
utilized on a more sequential basis throughout the year (Stier and Mildenstein 2001). 



23 

Female P. vampyrus give birth synchronously during a single annual peak, 
although the peak varies geographically and seasonally. Generally, females give birth 
between March and May to a single offspring, but twinning does occur (Medway 1969, 
Mickelburgh 1992). The gestation period is approximately 180 days (Kunz and Jones 
2000). Young bats are carried by their mothers for the first few days, but later are left at 
the roosts while their mothers forage. Young suckle from their mothers for 2-3 months 
(Lekagul and McNeely 1977). 

Camps of P. vampyrus in the Philippines once contained up to 100,000 bats 
(Mickleburgh et al. 1992, Nowak 1999). Unregulated hunting and habitat loss are the 
primary reasons for the decline in abundance for P. vampyrus (Mohd-Azlan et al. 2001). 
In many areas, this species is considered a nuisance because it feeds in fruit orchards 
(Medway 1969) and/or because of their noisy and conspicuous roosts (Kunz and Jones 
2000). P. vampyrus is also hunted for local consumption and controls on hunting are 
considered unenforceable. As a result, their numbers have declined severely (Heideman 
and Heaney 1989). P. vampyrus is also threatened by the rapid loss and degradation of 
mangroves for coastal reclamation and aquaculture, and by commercial logging and land 
clearing for palm/rubber estates (Heideman and Heaney 1989). The Malayan flying fox is 
listed as a species that that may become threatened with extinction if trade is not 
regulated (Brautigam 1992) 









CHAPTER 3 

FRUIT CHOICE AND CALCIUM BLOCK USE BY TONGAN FRUIT BATS: DO 

FRUIT BATS SEEK OUT CALCKJM IN THEIR DIET? 

Introduction 

Diets of wild animals can be low in essential nutrients. When minerals are 
deficient in the diet, animals often seek out concentrated sources of these nutrients. These 
sources may include natural mineral licks and foods that are rich in the deficient mineral 
(Klaus et al. 1998). High sodium concentrations often attract wild animals to mineral 
licks (Belovsky and Jordan 1981, Moe 1993, Tracy and McNaughton 1995), although 
minerals such as calcium and magnesium may be equally important (Jones and Hanson 
1985, Holl and Bleich 1987). When consumed by animals, these concentrated sources of 
minerals may help compensate for mineral deficiencies in the diet (Klaus et al. 1998). 

The idea that animals preferentially select nutrient-rich foods or a nutritionally 
balanced diet from among a broad array or foods is coined "nutritional wisdom" and is 
highly controversial. When given a choice of diets, animals often chose foods with the 
highest nutrient content or minerals in which they were deficient (Ozanne and Howes 
1971, Batzli and Pitelka 1983, Bromage and DeLuca 1984, Barclay 2002). However, 
other studies have found that animals failed to select foods that either met their dietary 
requirements or corrected for their nutritional deficiencies, but instead preferred palatable 
but nutritionally poor diets (Arnold 1964, Coppock et al. 1972, Muller et al. 1977, 
Oftedal and Allen 1996, Dierenfeld and McCann 1999, Zervas et al. 2001). 



24 



25 

Selection for concentrated sources of minerals may be associated with the sex of 
the consuming animal. For example, adult females of many species were more likely to 
use mineral licks, especially during pregnancy and lactation (Faber et al. 1993, 
Montenegro 1998). It has been postulated that reproduction in bats may be limited by 
calcium deficiency (Barclay 1994,1995). To provide supplemental calcium to their diet, 
fruit bats on occasion seek out calcium-rich fruits and leaves (Barclay 2002). 

This study was designed to explore whether wild Tongan fruit bats {Pteropus 
tonganus) sought out and preferentially consumed calcium when it was made available. 
Captive bats were tested as follows: (1) by documenting preference or avoidance of 
calcium-rich fruits and (2) by documenting use of commercial calcium blocks. It is 
hypothesized that bats would seek out calcium in their diet by consuming calcium-rich 
fruits and using the calcium blocks. 

Methods 

This study was conducted from December 2000 to August 2001 on the island of 
Tutuila, American Samoa (14° S, 170° W) in the South Pacific Ocean. The Tongan flying 
fox, P. tonganus, a medium-sized fruit bat (300-600 g), was used for this study. P. 
tonganus is common throughout the South Pacific (Miller and Wilson 1997) and is a 
feeding generalist that forages in both native forest and agricultural areas (Banack 1996). 

Tongan fruit bats (13 males, 8 non-reproductive females, 2 (lactating) 
reproductive females) were captured in mist nets and transported to a 4 x 3 meter 
screened outdoor structure (the "bat house") to allow for movement and limited flight by 
the captive bats. Following a two-day acclimation period, bats were individually tested 
for five days, for a total of seven days in captivity. Bats were given twice their body mass 
in food nightly (wet weight) of a high and low calcium fruit suspended from plastic cable 



26 

ties on a large wooden dowel rod (the food bar). The types of fruit given to the bats 
varied every day depending on fruit availability on the island. The high-calcium fruits 
used in the experiments were 2-3 times higher in calcium than the low-calcium fruits 
(1.06-12.28 mg/g, 5.75 ± 1.39 mg/g for high-calcium fruits, 0.55-2.46 mg/g, 1.30 ± 0.59 
mg/g for low-calcium fruits). Mineral concentrations for all native and agricultural fruits 
used in this study were based on previous mineral analyses (Nelson et al. 2000a). All fruit 
given to the bats were preferred foods of P. tonganus in American Samoa (Banack 1996). 
The most commonly used combination of fruits was local bananas and papaya, because 
they were readily available at local markets when native fruit species were unavailable. In 
addition to the high-calcium and low-calcium food types, a calcium block was suspended 
nightly from the food bar. The position of the calcium block was randomly assigned each 
night. The calcium block consisted of calcium sulfate and ground limestone, and 
contained 21-26% calcium (8 inl Pet Products, Inc., Fairport Harbor, OH). Water and a 
salt lick were available ad libitum. 

An infrared video camera (Sony Digital Handycam DCR-TRV 120) was placed in 
the bat house in front of the food bar each night to record fruit choices by the bat. The 
videotape recorded the first 1.5 hours of each nightly feeding session. Use of the calcium 
block was documented for each bat, as was the sequence of fruit choices. The first five 
fruit choices made by the bat were considered an indicator of fruit preference. The 
number of high-calcium fruits and low-calcium fruits chosen from the first five choices 
were analyzed statistically to document if bats sought out calcium in their diet. A 
binomial test was used to analyze preference for high or low calcium fruits (Hollander 
and Wolfe 1999). 



27 

To better understand preferences for fruit choice, I incorporated the effect of 
sugar and analyzed the relationship between calcium and sugar content using the 
multivariate technique Conjoint Analysis (Hair et al.1998). Conjoint analysis predicted 
bat choice for fruits when bats were given a subset of all fruits found on the island (see 
Green and Srinivasan 1978). Differences in sugar concentrations between samples of 
native and agricultural fruit were tested using a Mann- Whitney rank sum test (Sokal and 
Rohlf 1995). Sugar concentration of fruits was considered high if it was over 10%, 
intermediate if it was between 5-10% and low if it was less than 5% sugar. A sugar 
refractometer (Model # 300010, Sper Scientific, Scottsdale, AZ) was used to determine 
sugar values of different fruit types. High, medium and low calcium values were 
determined from previous mineral analysis of fruits (Nelson et al. 2000a). Both the sugar 
concentration and the calcium content of fruits were evaluated in Conjoint analysis to 
form part-worth estimates that were summed to totals. The largest part-worth totals 
resulted in high rankings for fruit preference. 

Results 
Choice of High-Calcium or Low-Calcium Fruits 

Results of 63 trials and 146 h of video were analyzed to test whether bats 
preferred high or low calcium fruits when given a choice among them. The first five 
choices made by each bat in a trial were documented, resulting in a total of 262 choices. 
Low calcium fruits were chosen 200/262 times, and preferred 76% of the time. Low- 
calcium fruits were highly preferred over high-calcium fruits (p < 0.001). Fifty-five of the 
62 times that high-calcium fruits were chosen, bats chose papaya. When papaya was 
removed from the high calcium fruit choice set, then low-calcium fruits were preferred 



28 

97% of the time (p < 0.001). Native fruit was chosen in the first five choices only once in 
262 trials, by a bat that took a single small bite and then did not choose it again. 

A native fruit (a fig, Ficus tinctoria) and an agricultural fruit (papaya, Carica 
papaya) were compared for sugar concentration using a Brix sugar refractometer. Papaya 
samples (n=12) averaged 12.6% sugar and were significantly higher in sugar (p = 0.001) 
than fig samples (n=15), which averaged 2.4% sugar. Although they provide an excellent 
source of calcium (O'Brien et al. 1998), figs were never consumed by bats and were 
highly avoided in this study. Despite using fresh, frozen, and three different species of 
figs (Ficus tinctoria, F. unirauniculata, F. scabra), figs were never eaten by the bats. 

Results of a Conjoint analysis indicated the sugar content of fruit was the basis for 
fruit preference and selection by all of the bats that were tested (Table 3-1). Fruits that 
were the most preferred were high in sugar and low in calcium, and the least preferred 
fruits were low in sugar and high in calcium. Even if the sugar was high, high-calcium 
fruits were still avoided by bats. I then reanalyzed the data to evaluate female choice 
among fruits (Table 3-2). The results were very similar to those for the entire data set, 
indicating that individual females did not forage differently from the group, they avoided 
high-calcium fruits and preferred high sugar fruit. Sample sizes were too small to perform 
a Conjoint analysis for reproductive females. 
Use of the Calcium Blocks 

Twice as many females used the calcium blocks as did males (4 females, 2 
males), and almost half (40%) of the females used the calcium blocks, including both 
reproductive females (Figure 3-1). Use of the calcium blocks by males was limited; only 
15% of the males (2 subadults) used the blocks. 



29 


Table 3-1. Conjoint analysis results for all bats showing part-worth estimate totals and the 


resultant rankings of fruit characteristics to determine fruit preference by P. tonganus. 


Sugar 


Calcium 




. Part worth 
Level 

estimate 


. Part worth 
Level 

estimate 


Total Ranking 


High 0.27 


Low 0.38 


0.65 1 


Medium -0. 1 


Low 0.38 


0.28 2 


High 0.27 


Medium -0.05 


0.22 3 


Medium -0. 1 


Medium -0.05 


-0.15 4 


High 0.27 


High -1.57 


-1.30 5 


Low -1.87 


Low 0.38 


-1.49 6 


Medium -0.1 


High -1.57 


-1.67 7 


Low -1.87 


Medium -0.05 


-1.92 8 


Low -1.87 


High -1.57 


-3.44 9 


Table 3-2. Conjoint analysis results for all females showing part-worth estimate totals 


and the resultant rankings of fruit characteristics to determine fruit preference by P. 


tonganus.. 


Sugar 


Calcium 




T . Part worth 
Level 

estimate 


, Part worth 
Level 

estimate 


Total Ranking 


High 0.27 


Low 0.56 


0.83 1 


Medium -0.09 


Low 0.56 


0.47 2 


High 0.27 


Medium -0.27 


0.00 3 


Medium -0.09 


Medium -0.27 


-0.36 4 


High 0.27 


High -1.41 


-1.14 5 


Low -1.88 


Low 0.56 


-1.32 6 


Medium -0.09 


High -1.41 


-1.50 7 


Low -1.88 


Medium -0.27 


-2.15 8 


Low -1.88 


High -1.41 


-3.29 9 





30 

Frequency of use was calculated as the number of times bats used the calcium lick 
divided by how many nights it was available to them. Frequency of use by males and 
females for the calcium blocks was very similar and did not exceed 10% (Table 3-3). 
Reproductive females used the calcium blocks with approximately four times the 
frequency (25%) of males or non-reproductive females. Most of the bats that used the 
calcium block were either reproductive females or subadults. In some cases, they used the 
calcium block before they ever chose fruit, and returned to use the calcium block 
intermittently while consuming fruit. 



a 








s 


100 - 






u 








a 


90 - 






u 








OX 


80 - 






*35 


70 - 






s 


60 - 






a u 

5 o 


50 - 






* -a 








•a jb 


40 - 






o 










30 - 






©^ 


20 • 






+rf 




_^^___ 




a 


10 - 


1 




O) 











- 


l 




C 




I 


S 








— 




Males (n=13) 


Non-i 




Non-reproductive Reproductive females 
females (n=10) (n=2) 

Figure 3-1 . Calcium block use by male, female and reproductive female Tongan fruit 
bats. Thirteen males, ten females and two reproductive females were tested. 



Table 3-3. Frequency of calcium block use by Tongan fruit bats. 



Group 



Males 

Non-reproductive 

females 

Reproductive 

females 



Number of Total number of Total 
bats in group trials mineral days of 

used trial 



13 
8 



3 
3 



47 
42 

8 



Frequency of use 
in the trials 

6% 
7% 

25% 



31 

Discussion 

Tongan fruit bats did not consistently seek out concentrated sources of calcium by 
preferring calcium-rich fruits or by using commercial calcium blocks with high 
frequency. Instead, the bats in this study preferred fruits that were high in sugar and low 
in calcium. If high-calcium fruits were chosen, the fruit was usually papaya, a preferred 
fruit that is high in sugar. Females used the calcium blocks more often than males. 
Reproductive females used the calcium blocks at more than four times the frequency of 
non-reproductive females and males. Subadult males were the only males to use the 
calcium blocks. 

Mammals often select foods to maximize their intake of carbohydrates like sugar 
(Provenza et al. 1996, Kimball et al. 1998), and fruit bats are no exception. Fruits high in 
sugars are highly preferred because they represent an important energy source for 
frugivorous bats (Dierenfeld and Seyjagat 2000b). Fruit bats prefer fruits such as papaya 
that are soft, succulent, and high in sugar, and will choose them over fruits low in sugar 
(Courts and Feistner 2000). Sugar, rather than calcium, appears to motivate dietary 
selection for fruits despite the importance of minerals to bat reproduction (Barclay 1995). 

There may be a temporal component to resource use by bats. In other studies, 
high-energy fruits were consumed first by hungry and dehydrated bats emerging from the 
day roost, followed by consumption of mineral -rich leaves later in the night (Kurta et al. 
1989, Elangovan et al. 2001). Tongan bats in this study had food available to them for ten 
hours a night, but only the first 1.5 h of feeding were recorded due to limited battery 
power and a lack of electricity in the bat house. This may have biased data collection 
toward documenting foraging for high-energy foods rather than calcium-rich foods. Thus, 
these results may not reflect feeding to consume deficient minerals that occurred later in 



32 

the night. Subsequent studies should record bat feeding at all times of the night to see if 
there is a temporal component to resource selection. 

It was hypothesized that bats would use concentrated calcium sources such as 
calcium blocks to relieve mineral deficiencies. Overall, frequency of use was low for the 
calcium blocks, but noteworthy gender differences emerged. The only male bats to use 
the calcium blocks were two subadults. They licked the calcium blocks in almost half of 
their trials, and sampled high-calcium native fruits that were generally ignored by other 
bats. Subadult males may experience calcium deficiency due to rapid growth, and may 
ingest supplemental calcium to relieve temporary deficiencies. 

Females, particularly reproductive females, used the calcium blocks in greater 
numbers and with greater frequency than did males. Twice as many females as males 
used the blocks, including use among all reproductive females tested. The two 
reproductive females used the calcium blocks at four times the frequency of either the 
males or non-reproductive females, and often used the calcium blocks before consuming 
any fruit. One of the lactating females removed high-calcium leaves from the 
Callophyllum neo-ebuticum (Clusaceae) tree in the cage and was seen consuming them at 
the food bar within 20 minutes of receiving sugar-rich agricultural fruits. Leaf 
consumption was also very high among reproductive females, resulting in supplemental 
calcium ingestion (Chapter 4). These results seem to indicate that in some cases, 
reproductive females may prioritize calcium ingestion over the ingestion of high-energy 
foods. This may indicate calcium deficiency among reproductive females. Unfortunately, 
very few reproductive females were caught, and their small sample size limits potential 



33 

interpretation of these data. Future work should examine food choice using larger 
numbers of reproductive females. 

It is difficult to assess if animals in this study foraged with "nutritional wisdom." 
The assumption of this work was that calcium was the most deficient component of the 
diet, and would be pursued first by bats when foraging. Instead, the majority of P. 
tonganus fed on high-sugar, high-energy, agricultural fruits soon after their presentation, 
potentially to maximize energy consumption. Although female Tongan bats were 
potentially deficient in calcium (Barclay 1995), they did not choose high-calcium fruits 
from among those offered. A limited number of bats did seek out minerals by using the 
calcium blocks. Pregnant females and rapidly-growing subadult bats used the calcium 
blocks the most, and were the most likely candidates for calcium deficiency. 

Factors that motivate fruit selection among Tongan fruit bats warrant further 
research, with experiments that include a larger number of reproductive females, and 
observations of feeding done at different times throughout the night. This may create a 
more complete picture of nutritional priorities and how they affect temporal patterns of 
resource use. 









CHAPTER 4 
FOLIVORY IN FRUIT BATS: ARE LEAVES A NATURAL CALCIUM 

SUPPLEMENT? 

Introduction 

Folivory, or leaf-eating by bats, is a well documented phenomenon (Marshall 
1985, Lowry 1989, Funakoshi et al. 1993, Kunz and Ingalls 1994, Kunz and Diaz 1995, 
Banack 1996, Tan et al. 1998, Ruby et al. 2000). Leaves are an important dietary source 
of minerals, carbohydrates, and protein, and are especially rich in calcium (Tan et al 
1998, Nelson et al. 2000b, Ruby et al. 2000,). Leaves are a consistent food source for 
bats; they are available year-round and are predictable in time and space (Kunz and 
Ingalls 1994, Rajan et al. 1999). Thus, leaves may provide a greater net return per 
foraging bout than ingestion of large amounts of low-protein fruit or the active pursuit of 
insects (Thomas 1984, Kunz and Ingalls 1994, Tan et al. 1998). In addition, steroid 
hormones found in leaves may influence bat reproductive activity (Wickler and Seibt 
1964, Kunz and Diaz 1995). 

Bats consume leaves by leaf- fractionation. This process includes masticating the 
leaves into a bolus, swallowing the liquid portion, and ejecting the flattened fibrous pellet 
(Lowry 1989, Funakoshi et al. 1993, Kunz and Ingalls 1994, Kunz and Diaz 1995). By 
rejecting the fibrous portion, bats are able to consume leaf nutrients without altering their 
digestive tract or increasing wing loading (Kunz and Ingalls 1994). Frugivorous bats 
appear to be pre-adapted for folivory by leaf fractionation; their dentition and gut 
morphology are specialized for extracting and digesting a largely liquid diet (Tedman and 



34 



35 

Hall 1985, Kunz and Ingalls 1994). To shift their diet alternately between one of fruits to 
leaves would involve little, if any, change in form of function of the gut or dentition 
(Kunz and Diaz 1994). 

Folivory was once thought to be rare among fruit bats, with leaves taken only 
when other food sources were scarce (Marshall 1985, Funakoshi et al. 1993, Pierson et al. 
1996). However, recent studies have shown that leaf-eating is both common and 
widespread among Old World flying foxes (Banack 1996, Tan et al. 1998, Ruby et al. 
2000). Folivory has been reported for at least 17 species of Old World Megachiroptera, 
and leaves eaten by bats include 44 species of plants represented by 23 families (Kunz 
and Diaz 1995). Bats locally consume a large variety of leaves. For example, Cynopterus 
brachyotis fed regularly on the leaves of 14 plant species in southern India (Tan et al. 
1998), and Pteropus dasymallus on nine species in Taiwan (Funakoshi et al 1993). The 
incidence of leaf pellets under feeding roosts in Taiwan was 37-50%, and occurred 
almost throughout the year (Funakoshi et al 1993). However, this may be an 
underestimate. Leaf pellets are often not noticed because they are inconspicuous among 
other plant material on the forest floor (Kunz and Ingalls 1994). 

Calcium is of particular interest in bat biology (Barclay 1994,1995, Kunz et al. 
1995, Bernard and Davison 1996). It has been proposed that females may be stressed for 
calcium due to the mineral demands of both pregnancy and lactation (Barclay 
1994,1995). To compensate for the large size of their offspring, bats donate their own 
skeletal calcium to build the bones of their young (Barclay 1995). Leaves represent a rich 
and consistent source of calcium to bats that are mineral stressed. Calcium concentrations 
are often much higher in leaves than in fruit (Nelson et al. 2000b, Ruby et al. 2000). 



36 

While some fruits may be high in calcium, it is not readily available if the Ca:P ratio is 
less than the optimum of 2 to 1 (McDowell 1992, Robbins 1993). The Ca:P ratio is three 
times higher in leaves than in fruits, which further suggests that leaves may be valuable 
for their high calcium content (Kunz and Diaz 1995, Ruby et al. 2000). 

This study examined if folivory among a sample of captive, wild-caught Tongan 
fruit bats (Pteropus tonganus). This study is the first to examine the amount of leaves 
that are consumed by individual fruit bats in a single night, and to calculate how much 
calcium folivory contributes to total daily calcium intake. I also describe gender and age 
differences in leaf consumption and explain an observed pattern of leaf eating. 

Methods 

Research was conducted from December 2000 to August 2001 on the island of 
Tutuila, American Samoa (14° S, 170° W) in the South Pacific Ocean. All 23 (13 male, 
10 female) Tongan fruit bats (Pteropus tonganus) were caught using large mist nets and 
transported to the "bat house." The bats consisted of four adult male, nine juvenile male, 
four adult female, and six juvenile female bats. Two of the adult females were lactating. 
The bat house was a 4 x 3 m outdoor wooden structure with an adjoining 4 x 3 m 
screened outdoor pen specifically built to house bats for these experiments. The outdoor 
pen contained a single Callophyllum neo-ebudicum (Clusiaceae) tree for roosting and leaf 
consumption (Trail 1994, Whistler 1994). This tree was the only leaf source for the bats 
in the present study. The bats could fly and move easily within the outdoor enclosure. 
Each night, bats were offered twice their body mass in confirmed bat foods from the 
island (Banack 1996). The fruit type varied each day depending on fruit availability on 
the island. See Chapter 3 for further details. Salt rings comprised of salt and mineral oil 
and contained 96-99 % salt (Pet Products, Inc., Hauppauge, N.Y.) as well as collected 



37 

rainwater were available to the bats ad libitum. Feeding trials were conducted on 
individual bats and lasted three to five days following a two day acclimation period, for a 
total of seven days in captivity. Only one bat was present and tested at a time in the 
outdoor enclosure. Fruit traps, or raised screen platforms, covered the ground of the 
enclosure to catch food and leaves dropped by the bats while they were feeding. 

The number of leaves and percentage of leaf eaten were recorded for each bat 
daily. Limited samples of representative leaves were collected and dried at 105°C for 24 
h. Leaf samples included both whole leaves and leaves partially eaten by bats. Samples 
were analyzed at the University of Florida in Gainesville, Florida USA. Dried samples 
were prepared and digested according to Miles et al. (2000). Calcium concentrations 
(ppm) were assessed by atomic absorption spectrophotometry (Perkin-Elmer AAS 5000 
Norwalk, CT.). All values were calculated on a dry matter basis. 

Males and females, and juveniles and adults, were compared to identify 
differences in leaf-eating behavior. The bats seemed to fall into three general categories 
from visual observation of Table 4-1 . Bats that habitually consumed leaves more than 
50% of the days they were in the pen were classified as habitual leaf-eaters. Some of the 
bats consumed leaves occasionally, less than half of the days that they were in the 
experiments, and were classified at occasional leaf-eaters. Some bats never consumed 
leaves and were therefore classified as non-leaf eaters. These three groups were later 
compared for total calcium consumption. 

Supplementary calcium values for each group were calculated as the total amount 
of leaf matter eaten by a bat multiplied by the calcium (Ca) concentration of the leaves 
(8861.47 mg/g Ca). To determine how much calcium folivory contributes to total daily 



38 



Table 4-1. Age and gender of bats used in the experiment, and number of leaves, amount 
of total leaf matter eaten (g), calculated calcium supplement (mg) gained by leaf-eating. 



Gender Age 



Leaf 
eater? 



Days 

leaves 

eaten (%) 



Total 
leaves 

eaten 



Total leaf 
matter eaten (g) 



Calcium 
supplement (mg) 



Male 

Female 

Female 

Female 

Male 

Male 

Male 

Female 

Female 

Male 

Male 

Male 

Female 

Male 

Female 

Male 

Male 

Male 

Female 

Male 

Female 

Male 

Female 



Adult 

Juvenile 

Adult 

Adult 

Juvenile 

Juvenile 

Juvenile 

Juvenile 

Juvenile 

Juvenile 

Juvenile 

Juvenile 

Adult 

Adult 

Adult 

Juvenile 

Adult 

Juvenile 

Juvenile 

Adult 

Juvenile 

Juvenile 

Juvenile 



yes 
yes 
yes 
yes 
yes 
yes 
yes 
yes 
yes 
yes 
yes 
yes 
yes 
yes 
yes 
yes 
yes 
yes 
yes 
no 
no 
no 
no 



50 
20 
100 
100 
100 
100 
33 
25 
40 
100 
60 
60 
67 
100 
67 
80 
50 
50 
25 







5 

0.75 

4 

7 

26 

12 

1 

2 

3 

8 

12 

10 

2 

14 

4 

11 

2 

4 

1 











1.45 


12.85 


0.3 


2.66 


1.07 


9.47 


1.82 


16.13 


6.76 


59.91 


3.12 


27.65 


0.26 


2.31 


0.52 


4.61 


0.81 


7.18 


2.08 


18.43 


3.12 


27.65 


2.6 


23.04 


0.55 


4.87 


3.64 


32.26 


1.04 


9.22 


2.86 


25.34 


0.52 


4.61 


1.07 


9.48 


0.26 


2.31 



























calcium intake, I calculated the average amount of supplemental calcium consumed for 
each of the three groups, and compared that to the average total calcium ingested by that 
group. Differences in consumption of leaves between gender and age were compared 
using two-tailed / tests. The Kolmogorov-Smirnov test was used to evaluate assumptions 
of normality for each variable, and Levene's test was used to evaluate the assumption of 



39 

normality between groups (Sokal and Rohlf 1995). Total calcium consumption for the 
three leaf-consumption groups were compared using a one way ANOVA. 

Results 
Ninety-four feeding trials were performed on 23 Tongan fruit bats. Leaves were 
consumed by 82.7% of the bats in this study. More males (92%) consumed leaves than 
females (70%). The total leaf mass eaten (g) differed (p = 0.02) between the sexes; males 
consumed an average of 9.55 + 6.99 SD leaves, and females consumed an average of 
2.97 + 2.04 leaves over the period of the feeding trial. This resulted in 22.14 + 16.00 SD 
(g) of additional calcium for males and 7.06 + 4.56 SD (g) for females. The amount of 
additional ingested calcium was different (p = 0.02) between males and females. Juvenile 
male and juvenile female bats differed (p = 0.04) in their consumption of leaves (24.22 + 
17.03 and 4.19 + 2.24, respectively); but male and female adult bats did not (p = 0.40). 
Twice as many juvenile males ate leaves than juvenile females (8 males: 4 females).The 
maximum number of leaves eaten in a single night (26) was by a young male. 



60 

a 



45 
40 
35 

e '60 

I J 30 

1 S 25 

3 S 

— •* 20 

c * 1S 

M 10 



& 



9 



I ■ 




Adult males 



Adult females 



Juvenile males Juvenile females 



Figure 4-1. Supplemental calcium ingested (mg/g) by folivory for adult and juvenile 
males and females. 



40 

The number of leaves eaten by individuals over a five-day period ranged from 
0.75 to 26 leaves. Overall, folivory provided 2.3 to 32.26 mg of additional calcium to the 
diet of fruit bats in the present study. There were similar numbers of habitual and 
occasional leaf eaters (11 and 8, respectively); however, habitual leaf eaters consumed 
significantly more calcium through leaf-eating than occasional leaf eaters (p = 0.04). 
Habitual leaf-eaters consumed an average of 10 + 6.56 SD leaves, which contributed an 
additional 23.89 + 15.05 SD mg/g of dietary calcium (Figure 4-2). This represented an 
average dietary increase in calcium of 1 1-46% when compared to the daily calcium 
consumption for each bat in that group. Occasional leaf-eaters consumed an average of 
2.34 + 1.54 SD leaves, which added an additional 5.75 + 3.82 g of calcium to their diet. 
Occasional leaf-eating represented an average dietary increase in calcium of 3-22% when 
compared to the daily calcium consumption for each bat. Non-leaf-eaters (n = 4) added 
no additional calcium to their diet. 

When eating leaves, P. tonganus often avoided the fibrous midrib, and instead ate 
around it (Figure 4-3). I chemically analyzed portions of a leaf cut along the midrib that 
did not contain the midrib to imitate leaf consumption by bats. These samples were 
compared to whole leaves that contained the fibrous midrib. The portions with the midrib 
contained 15.08 mg/g of calcium, and leaves without the veins contained 8.86 mg/g of 
calcium. Unfortunately, the work was largely exploratory and several leaves were used to 
produce only one analyzed sample for both the leaves containing the midrib and those 
that did not contain the midrib. Thus, a statistical analysis of this sample was not 
possible. 



41 



T3 

s 

E 

3 

c 
o 

£ 

|o 
a 
U 



Average daily calcium 
consumption 

Supplemental calcium 
from leaf-eating 




Habitual leaf-eaters 



Occasional leaf-eaters 



Non-leaf-eaters 



Figure 4-2. A comparison of total calcium ingested among habitual, occasional, and non- 
leaf-eaters. Total calcium is average daily calcium and supplemental calcium from leaf- 
eating combined. Total average daily calcium values for each group are calculated from 
values found in Chapter 5. 




Figure 4-3. Typical pattern of leaf consumption by Pteropus tonganus in American 
Samoa 

Discussion 

Previous studies of folivory (Marshall 1985, Lowry 1989, Funakoshi et al. 1993, 
Kunz and Ingalls 1994, Kunz and Diaz 1995, Banack 1996, Tan et al. 1998, Ruby et al. 



42 

2000) were based on indirect means of quantifying leaf-eating by bats and thus were 
unable to describe either the amount of leaves eaten per bat, or the sex and age of the leaf 
consumer. This study quantified both the amount of leaves consumed by individual bats 
and how much folivory contributed to total dietary calcium intake. The majority (83%) of 
wild-caught bats engaged in leaf-eating while in captivity. Both sexes consumed leaves, 
but male bats consumed more leaves than females, and juvenile males consumed both the 
greatest number and volume of leaves of all groups. Habitual leaf-eating bats could 
potentially increase their dietary calcium consumption by 46%. Clearly, folivory is both 
widely and frequently practiced by P. tonganus, and it has the potential to contribute 
significantly to the total amount of ingested calcium. 

The assumption in conducting this study was that the brevity of the time that bats 
spent in captivity would neither alter nor adversely affect the current mineral status of the 
bats used in the experiments. That is, the deficiencies or excesses of their native diet 
would influence their consumption patterns while in captivity. If this is true, then both 
male and female bats consume leaves as a regular part of their diet in the wild, but 
consumption patterns and volume of leaves eaten differs among sexes and groups. 

Some bats consistently consumed leaves each day of the experiment, while others 
only sampled the leaves intermittently. An interesting outcome of this analysis is that 
although the average daily (dietary) calcium ingestion was lower for habitual leaf-eaters 
than for occasional leaf-eaters, both groups were appeared similar in the total amount of 
calcium ingested. Habitual leaf-eaters exhibited the greatest variation in daily dietary 
calcium ingestion. Non-leaf eaters ingested the least calcium, but calcium ingestion by 



43 

non-leaf-eaters was very similar to the average daily calcium ingested by the habitual 
leaf-eaters. 

Kunz and Diaz (1995) observed only mature males carrying leaves, and 
hypothesized that folivory may be limited to adult male bats. Overall, adult males 
consumed significantly more leaves and ingested more leaf matter than adult females in 
this study. Interestingly, leaf-eating was practiced most often by juvenile males. Other 
work has suggested that compounds extracted by leaf fractionation could be possible 
regulators of reproductive activity for bats (Kunz and Ingalls 1994, Kunz and Diaz 1995). 
Erythrina leaves may contain one or more metabolites (alkaloids) important for 
reproduction, and they are consumed by P. tonganus on the island of Tonga (Harris and 
Baker 1959). Perhaps, in addition to being a rich calcium source that supports rapid 
growth, leaves may also influence reproductive activity in young male bats. 

Female bats consume leaves to access nutrients and minerals not available in 
fruits (Kunz and Diaz 1995). Leaves analyzed from American Samoa were very rich 
sources of calcium and other macrominerals. Leaves contained concentrations of many 
nutrients comparable or higher than those of ripe fruit preferred by fruit bats in Samoa 
(Nelson et al. 2000b). Leaves tended to have higher levels of calcium, sodium, 
manganese, and magnesium than ripe native or agricultural fruits (Nelson et al. 2000b). 
Also, leaves are widely available in both the wet and dry season (Whistler 1994). Thus, 
leaves may represent a rich, year-round, readily accessible source of concentrated 
minerals for female bats. 



44 

P. tonganus females in Samoa may also consume leaves for their high calcium 
content. Banack (1996) found that female Tongan fruit bats in American Samoa gave 
birth year-round, and young were seen on mothers all months of the year. She also 
observed copulations with pregnant females, suggesting that the female was nursing 
while allocating her own calcium for the skeletal formation of a new offspring (Barclay 
1995, Banack 1996). Both gestation and lactation are nutritionally demanding, and their 
combined effect may promote leaf-eating in females. In addition, the population of 
P. tonganus in American Samoa has increased 3 -fold over the last decade, following a 
series of three destructive hurricanes that decimated the bat population (Craig et al. 
1994b, Elmqvist et al. 1994, Brooke 1998). This population's rapid expansion may have 
resulted in additional calcium stress to reproductive females. However, despite this 
potential calcium stress, P. tonganus in American Samoa consistently chose low-calcium, 
high-sugar agricultural foods that resulted in inadequate calcium consumption, and high 
levels of retention that suggest calcium stress (Chapter 5). All together, the cumulative 
demands of gestation and lactation, overlapping generations, a rapid population increase, 
and a diet low in calcium, may promote in leaf-eating by female bats. 

Recent evidence indicates that leaves are consumed by bats throughout the year. 
There are reports of year-around leaf consumption based on either analysis of fecal 
remains, leaf parts discarded beneath roosts (Lowry 1989, Parry- Jones and Augee 1991a, 
Bhat 1994), or direct observation (Zortea and Mendes 1993). Bhat (1994) noted that leaf- 
eating was common in each month of the year in C. sphinx. Banack (1996) described 
year-round leaf use by both Pteropus samoensis and P. tonganus in American Samoa. 
This study is consistent with her findings; leaves were consumed in each of the eight 



45 

months of this study, with no apparent differences between months or seasons. However, 
the manner in which Tongan fruit bats consumed leaves suggests that they are actively 
avoiding the fibrous midrib, despite it being a rich source of calcium. Bats may avoid the 
midrib because it has high levels of tannins or secondary compounds (Dasilva 1994). 

This study has shown that males consumed significantly more leaves than 
females, and that folivory can contribute significantly to the total dietary calcium of leaf- 
eating bats. However, the motivation for folivory still cannot be ascribed to a single 
factor; both the high calcium content and presence of hormonal compounds in leaves may 
play integral roles. The propensity of males to consume large amounts of leaves suggests 
a hormonal motivation, but the calcium contribution to the diet due to folivory is 
significant and noteworthy. Future research should test whether hormonal compounds are 
present in the Callophyllum neo-ebudicum leaves and if these compounds influence 
reproductive cycles in this bat. Meanwhile, leaf-eating is a common practice among 
Tongan fruit bats, and the leaves provide a rich and concentrated calcium supplement to 
the often calcium-poor diet of fruit bats. 



CHAPTER 5 

BIOAVAILABILITY AND APPARENT ABSORPTION OF MTNERALS CONSUMED 

BY WILD TONGAN FLYING FOXES IN AMERICAN SAMOA 

Introduction 

Previous research on nutrition in bats has concluded that energy and protein are 
the most important dietary nutrients (Thomas 1984, Herbst 1986, Fleming 1988). 
Although dietary mineral composition and concentration are not often measures of 
dietary quality (Cole and Batzli 1979, Batzli 1986), recent studies have illustrated the 
importance of minerals in bat nutrition (Uhland et al. 1992, Barclay 1994,1995,). 
Nutrients that are consumed at marginal or inadequate levels with respect to requirements 
may limit animal performance (Oftedal 1991). Mineral nutrition can affect fecundity, 
number of litters, and survival of offspring (Batzli 1986, Delgiudice 1990) and are critical 
to the basic physiological functions of animals (McDowell 1992). Despite their 
importance to survival and fecundity, mineral requirements remain largely unknown for 
most species. 

Calcium is the most abundant mineral in the body, and one of interest in bat 
biology. For a pregnant or lactating female, inadequate calcium intake causes weakened 
bones and low milk production (McDowell 1992). For her dependent young, inadequate 
calcium from the mother results in inhibited growth, loss of body mass, and reduced 
mineralization of bone that can result in lameness and bone fractures to young bones 
(Radostits et al. 1994). Low dietary intakes also affect successive generations; offspring 
of rats fed a poor calcium diet survived but could not reproduce and had only 75-80% of 



46 



47 

the normal skeletal calcium content (Brommage and DeLuca 1984). The excessive 
mineral demands of pregnancy and lactation result in a negative calcium balance as 
females donate their own skeletal calcium reserves to build the skeletons of their young 
(Radostits et al. 1994, Bernard and Davison 1996). Raising several young in sequential 
years may result in osteoporosis in females, resulting in bones that fracture and easily 
break (Keeler and Studier 1992, Studier et al. 1994a). 

Requirements for calcium and all other minerals needed to maintain the health of 
bats are currently unknown. To examine mineral levels in bats, previous studies have 
used indirect methods to quantify minerals in the diet. Methods of calcium status 
evaluation have included analysis of blood plasma (Kunz and Stern 1995, Heard and 
Whittier 1997, Dierenfeld and Seyjagat 2000a, Kwiecinski et al. 2001), whole body 
mineral composition (Studier 1994, Studier and Kunz 1995), and fecal analysis without a 
knowledge of the types or quantities of foods consumed in the diet (Studier et al. 1991, 
Keeler and Studier 1992, Studier et al.l994b). However, none of these methods 
quantified the amount of minerals consumed, absorbed, or the absorption efficiency of 
minerals. Together, these factors can create a portrait of bat feeding and demonstrate the 
degree of mineral inadequacy in an individual. 

The present study was designed to determine mineral absorption efficiencies of 
flying foxes using apparent absorption. Apparent absorption has been used in a limited 
number of nutritional studies (Belovsky and Jordan 1981, Dierenfeld and Seyjagat 
2000b), but not on wild bats. This method measures both mineral intake and fecal 
excretion, and it can account for the unique manner of feeding by fruit bats. To feed on 
fruit and leaves, fruit bats chew the plant matter into a fibrous pellet, swallow the juice, 



48 

and expel the flattened pellet (Lowry 1989, Kunz and Ingalls 1994, Kunz and Diaz 1995). 
Previous studies were not able to account for minerals found in the expelled pellet. 

Due to a lack of mineral requirement standards for fruit bats, target levels for 
nutrients are based on estimated requirements reported for other mammals (NRC 1995), 
or what is fed in captivity to maintain breeding populations (Courts 1998). Previous 
research has compared values either to those of domestic laboratory mammals or to a 
generalized mammalian standard (Oftedal and Allen 1996, Dierenfeld and Seyjagat 
2000a). Mineral requirements established for rats and primates seem the most appropriate 
models to compare to fruit bats in the absence of true bat values. Rats are similar in size 
to bats and are monogastric, whereas primates are larger but share a monogastric gut, 
similar feeding habits, and perhaps an evolutionary past with bats (Pettigrew 1991). 

No studies have confirmed that bats are significantly different than other 
mammals in their general nutrition needs, but differences in absorption of minerals can 
occur among similar species, and dietary diversity is high within the family Pteropodidae 
( Walinski and Guggenheim 1974, Marshall 1985, Courts and Feistner 2000). Nutrient 
requirements are affected by such factors as growth rate, reproductive output, and 
metabolic needs (Oftedal and Allen 1996). These factors can be markedly different for 
rats and primates when compared to bats. Moreover, published requirements for 
laboratory animals are often in excess of true requirements to allow for ingredient 
variation and other margins of safety (Oftedal and Allen 1996). Thus, mineral 
requirements of domestic animals may be an inappropriate standard for fruit bats. 

The Tongan flying fox, Pteropus tonganus (Quoy and Gaimard, 1830) was used 
in this study. P. tonganus is one of the most widely distributed of all Pteropus species 






49 

(Koopman and Steadman 1995) and has adapted to many habitats and food types. It is a 
highly plastic forager and feeds on both native and agricultural fruits (Banack 1996, 
Pierson and Rainey 1992). Its wide geographic range and use of both native and 
agricultural fruits suggests a highly generalized digestive system. Additionally, the 
Tongan fruit bat is a mid-size fruit bat (300-600 g; Miller and Wilson 1997) so mineral 
absorption values are applicable to a wide range of body sizes. Overall, the Tongan fruit 
bat is an excellent study species because results should be widely applicable to many 
other flying fox species. 

This study used apparent absorption to quantify mineral ingestion and absorption 
in P. tonganus. Mineral absorption results were compared to both mammalian standards 
and other fruit bat mineral retention values to quantify if Tongan fruit bats met 
recommended mineral standards or were mineral stressed as determined by elevated 
absorption levels. This was the first attempt to quantify ingestion, absorption, and 
retention for minerals in wild populations of flying foxes. 

Methods 
Netting and Housing of Bats 

Research was conducted from September 2000 to August 2001 on the island of 
Tutuila, American Samoa (14° S, 170° W) in the South Pacific Ocean. Tongan fruit bats 
(13 males, 10 females) were captured in large mist nets (6-18 m, 4 inch mesh, Avinet, 
Inc.) attached to pulleys set high in coconut trees or on tall poles. Nets were raised at 
sunset and were checked every 30 minutes until midnight when they were taken down. 
Netting was conducted all over the island so that the bats used in the mineral retention 
trials would represent the bat population of the entire island. 



50 

Following capture, a single bat was transported to the "bat house," a 4 x 3 m 
wooden building with an adjoining 4 x 3 m screened outdoor structure. Both structures 
were built to temporarily house bats for these experiments. The outdoor structure was a 
wooden frame enclosed with rat wire and screen, and was lined with fishing nets to 
facilitate roosting and movement of bats. It contained a single Callophyllum neo- 
ebudicum (Clusiaceae) tree that could be used by bats for roosting and leaf consumption. 
The bats could fly and move easily within the outdoor structure. Water and salt licks were 
available to the bats ad libitum inside the outdoor structure at all times. Calcium blocks 
were occasionally available to bats as part of another experiment (Chapter 3). Salt licks 
consisted of salt and mineral oil, and calcium blocks consisted of calcium sulfate and 
ground limestone (8 inl Pet Products, Inc., Fairport Harbor, OH). 
Mineral Metabolism Experiments 

Bats were sexed, weighed, measured, and examined for injuries upon capture. 
Following a two day acclimation period, the bats were tested for five days in the bat 
house. Each bat spent a total of seven days in captivity and only one bat was present and 
tested at a time in the outdoor structure. Bats were given twice their body mass in food 
nightly (wet weight) so that hunger would not result in atypical food choices. To test if 
Tongan fruit bats would choose high calcium fruits if they were available (Chapter 3), 
each bat was presented nightly with equal masses of one high-calcium fruit and one low- 
calcium fruit. The fruit type varied each day depending on fruit availability on the island. 
Native fruits were typically high in calcium and agricultural fruits were low in calcium 
(Nelson et al. 2000a). The high-calcium fruits used in the experiments were 2-3 times 
higher in calcium than the low-calcium fruits (1.06-12.28 mg/g, 5.75 + 1.39 mg/g for 
high-calcium fruits, 0.55-2.46 mg/g, 1.30 ± 0.59 mg/g for low-calcium fruits). Mineral 



51 

concentrations for all native and agricultural fruits were based on previous work 
analyzing the mineral concentrations of bat fruit in Samoa (Nelson et al. 2000a, 2000b). 
According to Banack, all fruits given to the bats were preferred P. tonganus foods in 
American Samoa (Banack 1996, 1998). 

During fruit preparation and handling, disposable plastic gloves were worn, and 
stainless steel or plastic utensils were utilized to avoid mineral contamination. Utensils 
were washed using 1 : 1 vinegar and distilled water solution. Fruits were cut into equal- 
sized cubes, weighed, and suspended on plastic cable ties from a large wooden dowel rod 
(the food bar). Desiccation factors were determined from sub-samples of both high and 
low-calcium fruit, handled and prepared in the exact same way as the fruits given to the 
bats, but placed in a separate small cage within the outdoor structure each night. The 
representative fruits were weighed the next morning to establish a desiccation factor for 
each fruit type that was later subtracted from the samples to yield an accurate estimated 
wet weight value. Fruit traps, or raised screen platforms, covered the ground of the 
outdoor structure to catch food and leaves dropped by the bats while they were feeding. 
The fruit traps were washed daily with collected rainwater. 

All samples were collected the following morning. Food remains were separated 
into two categories: uneaten food still hanging from the food bar or dropped to the fruit 
traps below (hereafter called oris), and food that had been chewed on and sucked of all its 
juice and then spit out as small flattened disks (called ejecta). Fecal matter, representative 
samples for both fruit types, and partially eaten leaves were also collected, separated, and 
weighed. A wet mass was recorded for each type of sample (orts, ejecta, fecal, 
representative fruit, and leaves) for each day. Seeds were removed from all fruits before 



52 

weighing. Samples were placed in a drying oven for 24 hours at 105° C and dried to a 
constant mass and reweighed. Urine was not collected separately because it is not used 
for apparent absorption (McDowell 1992). All samples were stored in plastic bags in 
airtight plastic containers containing desiccant. The containers were stored in an air- 
conditioned laboratory to prevent mold growth until laboratory analyses were done. 

Analysis of Samples 

Samples were analyzed at the Animal Nutrition Laboratory at the University of 
Florida in Gainesville, Florida. Dried samples were weighed and dry ashed at 550°C for 
12 hours. Samples were prepared and digested according to the procedures of Miles et al. 
(2001). Mineral concentrations of Ca, Cu, Fe, Mg, Fe, Mn and Zn were assessed using 
flame atomic absorption spectrophotometry using a Perkin-Elmer AAS 5000 (Perkin- 
Elmer 1980) after wet digestion in HCL and dilution in 1% lanthanum solution. 
Phosphorus was measured separately using a colorimetric assay (Harris and Popat 1 954). 
Samples were analyzed in duplicate if sample size allowed. Standard reference material 
(citrus leaves 1572, National Institute of Standards and Technology, Gaithersburg, MD.) 
was run with each sample set. 

Data analyses included calculating the total mineral consumption by each bat for 
each day. The mineral concentration numbers of fruit were multiplied by the amount of 
fruit that was consumed each day. This resulted in a value for total mineral intake. 

Apparent absorption was calculated using the following equations. Mineral 
calculations were calculated on a dry matter basis. Each calculation was performed for 
each mineral, for each bat, for each day they were housed in the outdoor structure, 
resulting in 805 apparent absorption values. The average absorption value for each bat, 
for each mineral, for all days in captivity was averaged and compared. 



53 

Total mineral intake (g) = 

(total amount of fruit offered (g)) - (total amount of orts (g)) - (total amount of 
ejecta (g)) 

Apparent mineral absorption (%) = 

(total mineral intake) - (fecal mineral) 
(total mineral intake) x 1 00 

Statistical Analysis 

To statistically analyze the data set, mineral apparent absorption values among all 
bats were tested for normality using the univariate Shapiro- Wilkes test. The values were 
not normally distributed, so the data were rank-transformed (Sokal and Rohlf 1995). The 
transformed data were analyzed using Principal Components Analysis (PCA) to evaluate 
patterns of variation between different bat sexes, ages, and reproductive states of the bats. 
PCA assesses relationships of independent variables within a single data set and places 
factors that are ecologically similar in close proximity in ordination space (Garigal et al. 
2000). Comparisons of apparent absorption values for each mineral between P. tonganus 
and the giant flying fox Pteropus vampyrus (Chapter 7) were analyzed using a two 
sample t-test assuming equal variance (Sokal and Rohlf 1995). 

Results 
Mineral Consumption 

A total of 1 1 5 feeding trials were performed on 23 wild Tongan fruit bats. The 
bats consisted of four adult male, nine juvenile male, four adult female, and six juvenile 
female bats. Two of the adult females were lactating. When all values for average mineral 
absorption values were analyzed, PCA resulted in no clear patterns or trends for either the 
macrominerals or trace elements (Table 5-1). Subsequent t-tests failed to distinguish clear 
patterns among the data for different sexes, ages, and reproductive classes for all 



54 

minerals. Thus, data were grouped and evaluated as a single data set for all future 
analyses. 

Table 5-1. Results of the principal components analysis comparing average mineral 

absorption values for all bats. 

Factor 1 Factor 2 

Eigenvalue 3.33 1.26 

Percent variation explained 47.50 17.98 

Contributions of each individual variable 
Calcium 0.45 0.26 
Phosphorus 0.32 0.52 
Magnesium 0.34 0.14 
Zinc 0.29 -0.69 
Iron 0.28 0.24 
Manganese 0.48 -0.12 
Copper 042 -0.32 

The results of the PC A for mineral components identified two factors with 
eigenvalues > 1 that together explained 65.48% of the variation. Factor 1 loaded 
positively for all minerals, but the factor scores were highest for calcium, manganese and 
copper. Factor 2 loaded positively for all minerals but zinc, manganese, and copper. 
Phosphorus loaded most heavily and positively of all the minerals in Factor 2 (0.52), and 
calcium was the second highest with the factor score 0.26. 

Bats were offered an average of 566.7 + 109.30 g (wet weight) of food per day of 
which an average of 332.2 + 1 13.00 g was consumed. Tongan bats consumed 85% of 
their body mass daily on a wet matter basis and 17% on a dry matter basis. Despite the 
large amount of food rejected daily, both the amount of food offered and the amount 
consumed consistently failed to meet the required mineral standards established for 
laboratory animals (Table 5-2). Calcium (Ca), phosphorus (P), manganese (Mn), and 
copper (Cu) were below the required levels for both the diet offered and consumed. Zinc 



55 

(Zn) greatly exceeded the rat and primate requirements in both what was offered and 
consumed. The amount of magnesium (Mg) offered was sufficient, but it was not 
consumed in adequate amounts to meet requirements. Both the amount of iron (Fe) 
offered and consumed met the requirements for the rat but not for the primate. 



Table 5-2. A comparison of nutrient levels offered and consumed by wild P. tonganus to 
values for standard diets of rats and primates.. 





Ca 


P 


Mg 


Zn 


Fe 


Mn 


Cu 


Rat 1 


0.56 


0.33 


0.06 


11 


39 


11 


6 


Primate 2 


0.54 


0.43 


0.16 


11 


196 


30 


10 


Diet offered to bats 


0.23 


0.23 


0.21 


54 


100 


9 


5 


Diet consumed by bats 


0.07 


0.11 


0.07 


50 


69 


4 


2 


Diet consumed by bats 


0.08 


0.11 


0.08 


50 


70 


5 


2 


eating leaves* 

















1= NRC 1995, 2= Oftedal and Allen 1996. 

*Folivory data are from Chapter 5 

Zn, Fe, Mn, and Cu are reported as ppm, and values for Ca, P, and Mg are reported as 

percent 

The addition of minerals due to leaf-eating (Chapter 4) slightly raised ingested 
mineral levels but overall had little impact on mineral consumption for Tongan fruit bats. 
Folivory slightly raised Ca and Mg levels, but only to one-seventh of the calcium 
requirement for rats and one half the magnesium requirement of primates. 

Calcium and phosphorus interact and influence the absorption of each other. 
Calcium is best absorbed when it occurs at a Ca:P ratio of 1 :1 to 2:1 (McDowell 1992). In 
this study, the expected quantity of calcium as well as the Ca:P ratio of 1:1 were not 
achieved in the food consumed by P. tonganus. The diet offered contained a 1:1 ratio, 
but the Ca:P ratio for ingested food was 0.6:1; approximately half the expected 
requirement. A low ratio can inhibit calcium absorption making less available to the 
animal for physiological functions (McDowell 1992). 



56 

Additional minerals were available to bats from drinking water. Collected rain 
water was given daily to bats. The values for 10 ml of Samoan rainwater (n=2 samples) 
are reported in ppm: Ca = 0.1 1, P = 0.06, Mg = 0.09, Zn - 0.05, Fe = 0.0, Mn = 0.02, 
Cu = 0.02. The amount of water consumed by bats remains unknown, so the mineral 
contribution of water to the diet could not be calculated. Salt licks and calcium blocks 
may have contributed to mineral ingestion, but use and amount of minerals ingested 
could not be quantified. 
Mineral Absorption 

Results of the present study were compared to a parallel investigation on the 
captive giant flying foxes P. vampyrus (Table 5-3). The P. vampyrus study was 
performed on groups of captive bats (Chapter 7) while P. tonganus were tested 
individually, but all other techniques used were identical for both studies. The captive 
colony of P. vampyrus includes lactating females and their offspring as well as non- 
reproductive females. Mineral apparent absorption values for the 23 P. tonganus used in 
the present study and those of the three pens of P. vampyrus have been averaged and 
presented in Table 5-3 as a single value for each species and mineral. 

The apparent absorption values reported for both P. vampyrus and P. tonganus are 
similar. P. vampyrus met their mineral needs for both amount of minerals offered and 
consumed for all minerals but copper (Chapter 7). In contrast, P. tonganus was deficient 
in amounts consumed for Ca, P, Mn, and Cu (Table 5-2). There was a difference (p < 
0.05) in apparent absorption values between P. tonganus and P. vampyrus for calcium (p 
= 0.04), phosphorus (p = 0.001), zinc (p - 0.001) and iron (p = 0.001), where P. tonganus 
had higher absorption values for those minerals. 



57 



Table 5-3. A comparison of mineral apparent absorption values for two species of flying 
foxes, Ptero pus tonganus and Pteropus vampyrus*. 





Pteropus vampyrus 


Pteropus tonganus 


Calcium** 


65% 


79% 


Phosphorus** 


63% 


84% 


Magnesium 


68% 


73% 


Zinc** 


39% 


85% 


Iron** 


63% 


76% 


Manganese 


69% 


68% 


Copper 


72% 


75% 



*Data sets used identical analytical procedures. P. vampyrus information is found in 

Chapter 7. 

** significantly different (p < 0.05) absorption values between species 

Discussion 

This study showed that wild Tongan fruit bats have high mineral absorption 
values. The absorption values for P. tonganus are more similar to those found for P. 
vampyrus, another flying fox species, than to either rats or primates (see Table 5.4). 
However, when compared to P. vampyrus, the Tongan fruit bat had significantly higher 
absorption rates for several important minerals. This suggests that the Tongan fruit bat 
population may be highly mineral stressed for several macrominerals that are only 
available in minimal quantities in their diet. 

Because the apparent absorption technique has never been attempted before on 
wild Tongan fruit bats, it is a challenge to find values to compare to those found in the 
present study. Apparent absorption values for monogastric animals and another species of 
bat, Desmodus rotundus, are presented in Table 5-4 to serve as a basis of comparison to 
values found for P. tonganus in the present study. The values presented are for typical 
animals that are not under conditions of mineral stress. For the minerals being studied, 






58 



values for monogastric animals are typically low and do not exceed 50% absorption 
under normal conditions. 






Table 5-4. Mineral apparent absorption values for selected monogastrics species. 
Absorp tion values found in this table are from non-mineral stressed individuals- 
Mineral Animal Absorption (%) Reference 



Calcium Human 21% 

Fox squirrel 30-39% 
Vampire bat 16-24% 

Phosphorus Human 24-3 1 % 

Pig 17-47% 

Vampire bat 18-44% 

Magnesium Human 46% 

Pig 50-60% 

Vampire bat 25-29% 

Zinc Human 14% 

Rat 17-20% 

Manganese Human 3-4% 

Rat 

Iron Human 

Rat 



Copper 



3-4% 
22% 
9-60% 
Vampire bat 5-14% 
Human 25-70% 

Sheep 3-13% 



Coudrayetal. 1997 

Havera 1978 

Coen 2002 

Hevesy 1948 

Jungbloed and Kemme 1990 

Coen 2002 

Coudrayetal. 1997 

Miller 1980 

Coen 2002 

Coudrayetal. 1997 

Tidehagetal. 1988 

Hurley and Keen 1987 

Greenberg et al. 1943 

Coudrayetal. 1997 

Fairweather-Tait and Wright 1991 

Coen 2002 

Strickland et al. 1972 

Suttle 1991 



However, when dietary minerals are deficient in the diet, mineral absorption often 
increases due to homeostatic mechanisms that compensate for inadequate intake of that 
nutrient (Ammerman 1995). Values for normal, non-stressed animals are given as a 
reference for each mineral. Absorption values for mineral deficient animals are typically 
3-4 times higher than are found for animals with adequate mineral intake. 

Together, these two tables illustrate how elevated the apparent absorption values 
are for P. tonganus in this study. These values for P. tonganus are much higher than are 
typical of other monogastric animals under normal conditions, and are instead similar 
tothose found for animals that are nutritionally stressed for minerals. Mineral apparent 



59 

Table 5-5. Apparent absorption values for minerals in animals under nutritional stress 

compar ed to animals at normal mineral intake levels. 

Mineral Animal Absorption (%) Mineral Status Reference 

Calcium Human 28% normal intake Brine and Johnson 1955 

43% deficient intake Brine and Johnson 1955 

58% preterm infant Bronneret al. 1992 

75% growing child RDA 1989 

Phosphorous Human 24-31% adult Hevesy 1948 

7 1 % preterm infant Koo and Tsang 1 99 1 

Magnesium Rat 26% normal intake Brink et all 992 

57% deficient intake Brink etal. 1992 

Iron Human 2-15% normal adult Josephs 1958 

20-60% anemic adult Josephs 1958 

Manganese Rat 3-4% adult Greenberget al. 1943 

20% young Keen etal. 1986 

absorption values for P. tonganus are most like those off. vampyrus, but P. tonganus 
values are still significantly higher for several minerals. However, the majority of 
absorption values for P. vampyrus are those of lactating females and their pups. The 
highly elevated apparent absorption values for P. tonganus suggest that this population 
may be highly stressed for mineral nutrients while consuming its current diet in the wild. 
The diet of P. tonganus typically includes a large volume of low-nutrient agricultural 
fruits (Banack 1996), which potentially contributes to nutrient deficiency in this species. 
Bioavailability and Absorption 

Bioavailability is defined as the degree to which an ingested nutrient is absorbed 
in a form that can be used for metabolic functions by an animal (Ammerman 1995). Total 
intake of a nutrient depends on both the intake and bioavailability of the nutrient (Oftedal 
1991). The absorption level of a mineral provides an estimate of its bioavailability. For 
example, high absorption indicates that the mineral's bioavailability is high and low 
absorption indicates that the mineral's bioavailability is low. Minerals found in plants are 
often less bioavailable than in animal sources because fiber in plants binds to minerals 



60 

and makes them unavailable for absorption (Soares 1995). Because of this, only 30-50% 
of ingested calcium in humans is absorbed by the body (Arnaud and Sanchez 1996, 
Bronner 1998). 

Fruit bats consume foods by chewing them into a bolus, swallowing the liquid 
portion, and ejecting the flattened fibrous pellet (Lowry 1989, Funakoshi et al. 1993, 
Kunz and Ingalls 1994, Kunz and Diaz 1995). By rejecting the fibrous portion and 
swallowing the juice, fruit bats may be increasing the bioavailability of the minerals in 
fruits. Levels of minerals were usually more concentrated in the ejecta pellet than in the 
diet samples in this study. This suggests that the minerals are in a less soluble form than 
in fruit and remain in the fiber portion after it has been ejected. To test this hypothesis, 
apparent absorption was recalculated by including the fibrous pellet in the feces 
calculation, assuming the portion of calcium in the pellet would not have been absorbed 
in the gut. The calculation resulted in a 5-20% decrease in mineral apparent absorption 
values. Thus, the unique pattern of bat consumption may remove the highly absorbable 
minerals from the pellet by placing them in solution. Because calcium must be in solution 
to be absorbed (Bronner and Pansu 1999), apparent absorption will be high after 
ingesting the fruit juices because the ingested minerals are highly bioavailable (Pansu et 
al. 1993, Duflos et al. 1995,). This manner of feeding results in minerals that are readily 
absorbed, resulting in high absorption values. Bioavailability and solubility of minerals 
are crucial under conditions of low mineral intake (Bronner and Pansu 1999). 

In addition to their unique feeding behavior, fruit bats have several anatomical 
adaptations that further increase mineral absorption. The stomachs of fruit bats are large, 
and the small intestine is long and convoluted, both potential adaptations to increase 



61 



absorptive surface area (Dempsey 1999). In addition to the solubility of the mineral, time 
spent in the intestine is the differentiating factor as to how much mineral is absorbed 
(Bronner 1998). The relatively long intestine in Old World fruit bats may be up to nine 
times their body length (Okon 1977). The added surface area of the stomach and 
intestines may be essential to counteract short gut retention times among fruit bats (30 
minutes, Tedman and Hall 1985), and to increase nutrient absorption time of 
nutritionally-poor foods. 
Mineral Stress 

Fruit bats may have higher mineral absorption values than other monogastric 
animals due to their unique feeding patterns and the potentially high bioavailability of 
minerals in their food. However, high absorption values also suggest mineral stress. An 
animal will absorb more of a nutrient if the nutrient is deficient in the body or diet 
(Ammerman 1995). P. tonganus had higher absorption levels for many critical nutrients 
when compared to P. vampyrus, a flying fox twice its size (Kunz and Jones 2000). The 
highly elevated absorption values of P. tonganus are of particular interest because P. 
tonganus did not meet its expected mineral requirements for either offered or consumed 
food, whereas P. vampyrus met all of its expected nutrient requirements. 

Although agricultural fruits were nutrient poor, and Tongan fruit bats did not meet 
their expected mineral requirements, bats consistently rejected 41% the offered food and 
consumed only 85% of their body mass in food nightly. Unlike other bats that consume 
agricultural fruits, P. tonganus did not consume 2.5 times their body mass to meet its 
nutrient requirements (Dempsey 1999). Although low in nutrients, foods offered to P. 
tonganus were documented as preferred bat foods in American Samoa (Banack 1996, 
1998), and were highly preferred to native fruits by fruit bats in fruit preference 



62 

experiments (Chapter 3). In addition to being low in nutrients, the foods ingested resulted 
in a low Ca:P ratio of less than 1:1, which further inhibited mineral absorption 
(McDowell 1992). Tongan fruit bats consistently chose and consumed low-nutrient fruits 
that did not meet the mineral requirements of the rat, primate or mammalian standard. 

Tongan fruit bats in this study consumed only one-eighth of the expected 
requirements for calcium and one-third of the phosphorus requirements required for a rat. 
Yet despite not meeting these requirements, P. tonganus births have been observed year 
round in Samoa, with high numbers of pregnant and lactating females seen throughout 
the year (Banack 1996). The population has increased three- fold in the last decade 
following a series of hurricanes that had severely reduced the population (Craig et al. 
1994b, Pierson et al.1996, Brooke 1998). Thus, rapid population expansion coupled with 
the high nutrient cost of bearing a single young (Barclay 1994, Kunz and Stern 1995), 
and despite the consumption of nutrient-poor food, brings into question the source of 
minerals, particularly calcium, to support rapid population growth. 

The priority of all mammals is to maintain calcium concentrations in plasma close 
to 2.5 mmol (100 mg) l" 1 despite fluctuations in the amount of calcium ingested (Arnaud 
and Sanchez 1996, Hurwitz 1996). This concentration is needed to maintain calcium 
functions such as cellular metabolism, blood clotting, enzyme activation, and 
neuromuscular action (McDowell 1992, Soares 1995). Plasma levels will not reflect 
mineral deficiency for minerals such as calcium, because homeostatic mechanisms 
maintain calcium levels despite dietary deficiencies (McDowell 1992). Bone acts as a 
large storehouse for calcium: 99% of the calcium in the body is stored in bone. If plasma 
calcium concentrations begin to decrease, calcium is quickly mobilized and resorbed 



63 

from bone to return plasma calcium levels back to normal (Bronner 1992, Garel 1987). 

Bone calcium is in a constant state of flux, and it is resorption rather than accretion that is 

highly responsive in restoring plasma calcium levels (Kwiecinski et al. 1987a). 

Changes in bone calcium therefore reflect the extent to which dietary calcium 

meets the calcium requirements of bats (Bernard and Davison 1996). Females readily 

allocate their own skeletal calcium to build the skeletons of their offspring, and exhibit 

marked bone-thinning and structural changes to bone as a result of the calcium demands 

of pregnancy and lactation (Kwiecinski et al. 1987b, DeSantiago et al. 1999). Calcium 

levels and bone density can be restored following lactation, with the consumption of an 

adequate calcium diet (Kwiecinski et all 987b). However, Tongan fruit bat females in my 

feeding trials were not consuming a nutritionally adequate diet based on expected values. 

Their diet was marginal for most of the macronutrients examined in this study, and 

extremely poor in calcium. Thus, if and how Tongan fruit bats are able to rebuild their 

skeletons following pregnancy and lactation remains unknown. 

Future research 

The only true measure of female calcium levels is a bone density test. Porosity of 

bone serves as an indicator of the calcium status and could determine if the skeletons of 

P. tonganus females are porous and osteoporitic, or are healthy, having recovered from 

the demands of raising young. To test these ideas further, I predict the following 

concerning bone density. 

1 . Flying fox species that prefer agricultural fruits should have lower bone density 
than those that prefer nutrient-dense native forest fruits. This prediction could be 
tested using P. tonganus and its congener Pteropus samoensis in American Samoa 
(Banack 1996, Pierson and Rainey 1992, Nelson et al. 2000a) ox Pteropus 
vampyrus lanensis and Acerodon jubatus in the Phillipines (Steir and Mildenstein 
2001, Mildenstein 2002). 






64 



2. Tongan fruit bat's preference for nutrient-poor agricultural fruits is relatively 
recent and corresponds with human settlement of the South Pacific islands. This 
could be tested by conducting bone porosity tests on archaeological bat remains 
that predate human settlement of the islands and comparing them to the bones of 
modern Tongan fruit bats (see Steadman 1991, Kirch et. al. 1992). 

3. Female bats that consume agricultural fruits and have had many pups should have 
lower bone density and higher bone porosity than adult non-parous females or 
adult males of the same species that consume an agricultural diet. 

4. Supplementing the diet with calcium should lower mineral absorption rates and 
increase bone density. The diet of P. vampyrus in captivity was supplemented 
with additional Ca. Therefore, with Ca supplementation, P. tonganus should 
increase bone density and decrease mineral absorption over time, resulting in 
values similar to those of P. vampyrus. 

Theoretically, animals should evolve feeding behaviors that enhance the intake of 
limiting nutrients (Oftedal 1991). Fruit bats may have adapted behaviorally and 
anatomically to increasing the bioavailability and absorption of minerals from their diet. 
These adaptations probably arose in response to the evolutionary constraints on increased 
wing loading from fiber in fruits (Dudley and Vermeij 1992,1994) and to flying fox's 
preference for high-sugar, low-nutrient fruits (Parry- Jones and Augee 1991a, Nelson et 
al. 2000a). However, it is not yet known if the preference for high-sugar, low nutrient 
foods predates the arrival of humans and agriculture on South Pacific islands. 

In conclusion, by producing a rejected fiber pellet and swallowing only the juice 
while eating fruits and leaves, bats may place highly bioavailable minerals in solution. 
These minerals can then be readily absorbed in the abundant surface area of a large and 
highly convoluted stomach and small intestine. The elevated apparent absorption values 
found in P. tonganus suggest that ingested minerals are either highly bioavailable and 
readily absorbed, or this population is in severe mineral stress due to its preference for 
agricultural fruits. Future work may distinguish if elevated absorption values are unique 
to Tongan fruit bats or are typical of other flying fox populations. Much work remains to 



65 

be done on the mineral nutrition of fruit bats so that base values can be established for 
different species and animals in different physiological states. 



CHAPTER 6 

NUTRITIONAL LANDSCAPE ECOLOGY AND HABITAT USE BY TONGAN 

FLYING FOXES IN AMERICAN SAMOA 

Introduction 

Flying foxes of the genus Pteropus are strong fliers capable of traveling long 
distances (Nelson 1965, Eby 1991, Spencer et al. 1991, Banack 1996, Palmer and 
Woinarski 1999, Shilton et al. 1999, Palmer et al. 2000). Pteropus species may commute 
up to 50 km nightly, traveling at speeds of 40 km/h while searching and foraging among 
food patches throughout the landscape (Richards 1990, Spencer et al. 1991, Palmer and 
Woinarski 1999, Banack 1996). This allows flying foxes to access patchily distributed 
fruit, nectar, and flowers, and to avoid local food shortages by seeking out distant, 
scattered food resources (Bronstein 1995). Thus, an entire forest or all of a small oceanic 
island may represent potential foraging habitat to a flying fox. 

The selection of a foraging patch within a heterogeneous landscape has been 
described as a hierarchical decision process that occurs at different levels: regional, 
landscape, plant community, or at the level of the individual plant (Senft et al. 1987). 
Potentially using significant powers of spatial memory and learning, flying foxes range 
widely with information on both the location and quantity of resource patches within the 
landscape (see Lima and Zollner 1996, Zollner and Lima 1999). For example, the black 
flying fox (Pteropus alecto) exploited landscape patchiness at two scales, between broad 
vegetation types and within vegetation types, and selected sites that were rich in 
resources from among the homogeneous forest matrix (Palmer et al. 2000). Its selection 



66 



67 

highlighted the patchiness of resources in the landscape (Palmer et al. 2000) and 
demonstrated the bat's awareness of differences in habitat quality among the patches. 

Animals should utilize resource patches within the landscape in a way that 
maximizes fitness (Lima and Zollner 1996). In certain stages of a life cycle, nutritional 
needs can be very specific, and filling those needs can be crucial for population survival. 
Habitats in which such needs can be met are often considered key to population 
persistence (Kozakiewicz 1995). For example, calcium is often limited in the diet of bats, 
and females may be calcium stressed during pregnancy and lactation (Keeler and Studier 
1992, Radostits et al. 1994, Studier et al. 1994a, Bernard and Davison 1996). Inadequate 
levels of dietary calcium can result in low milk production in females and inhibit growth 
in offspring (McDowell 1992, Radostits et al. 1994). Limited calcium availability in the 
diet has the potential to limit fecundity and survivorship of animal young (Batzli 1986), 
which may affect population levels in bats (Barclay 1995). Increased calcium 
requirements during pregnancy and lactation may be relieved by the consumption of 
concentrated sources of calcium such as figs (Ficus spp.) or other calcium-rich fruits 
(Nelson et al. 2000a, Ruby et al. 2000). Foraging areas that contain concentrated sources 
of calcium should potentially be important for reproductive bat populations and for 
population persistence. 

Before the arrival of Polynesians more than 3,000 years ago, most of Tutuila, 
American Samoa, was covered in native rainforest (Cole et al.1988, Whistler 1992, Hunt 
and Kirch 1997). Since then, human activities such as land clearing and shifting 
cultivation have altered much of the forest area and replaced rainforest with mixed crops 
and residential areas (Cole et al. 1988). Previous nutritional analysis of native and 



68 

agricultural fruits indicate that native fruits are a much more concentrated source of 
nutrients, particularly for the mineral calcium (Nelson et al. 2000a). Thus, patches that 
contain native fruits may represent higher quality habitat than those that contain nutrient- 
poor agricultural fruits. Habitat patches that contain calcium-rich fruits could potentially 
be used to increase dietary calcium consumption. 

The Tongan fruit bat, P. tonganus, is a habitat and feeding generalist, that forages 
in both native and agricultural areas on 42 species of plants on Tutuila, American Samoa 
(Wilson and Engbring 1992, Trail 1994, Banack 1996), but the extent of use of each 
habitat type is unknown (Brooke 1998). Tongan flying foxes are able to transverse the 
length of Tutuila island in a single night, and use different sides of the island for foraging 
throughout the night (Banack 1 996). Tutuila is isolated from other islands by an ocean 
barrier of 100 km. Thus, the island represents a single, isolated, heterogeneous foraging 
area. This results in a unique situation that allows one to study the entire foraging area 
available to P. tonganus in American Samoa. 

Nutritional landscape ecology combines the concepts of nutritional ecology and 

landscape ecology to determine if animals select nutrient-rich areas as they forage within 

the landscape. In the present study, I evaluated whether Tongan fruit bats foraged 

preferentially in high calcium habitat types to increase consumption of calcium. It was 

predicted that Tongan flying foxes, particularly pregnant and/or lactating females, would 

forage within calcium-rich areas to obtain supplemental calcium in their diet. 

Methods 
Major Vegetation Types 

This study was conducted between February 2000 and August 2001 on Tutuila, 

the largest island in American Samoa (14° S, 170° W) in the South Pacific Ocean. The 



69 

three major vegetation types on present-day Tutuila are native forest, mixed agroforest, 
and village agricultural areas. Native forest is the climax forest for the island and includes 
upland, mangrove, moss, and coastal forest (Cole et al. 1988, Whistler 1994). Native 
forest on Tutuila includes a rich diversity of plant and tree species with a 30% level of 
endemism (Whistler 1992, 1994). Mixed agroforest results from disturbance, either 
natural or anthropogenic, and is a transitional stage between plantation land and native 
forest. In this forest type, fruit trees are planted among secondary growth forest trees 
(Cole et al. 1988). Village agricultural lands include land cleared to grow fruit trees in 
sparse density, and are adjacent to residential areas that include villages, plantations, and 
roads (Cole et al. 1988). A vegetation habitat map of Tutuila was available for Tutuila 
and coded for the three major vegetation types (Cole et al. 1988, Freifeld 1998). 
Nutritional Classification of the Major Vegetation Types 

The three vegetation types of the island were categorized as either high, 
intermediate, or low in nutrient availability based on the calcium content of the fruits 
found within each habitat (Table 6-1). Nutritional analysis indicated that minerals were 
more concentrated in native fruits than in agricultural fruits (Nelson et al. 2000a). Figs 
are an especially rich source of calcium and are found mostly in native forests (Whistler 
1992, Nelson et al. 2000a). Because native forests contain calcium-rich fruits and figs, 
these forests represented calcium-rich habitat. Mixed agroforest represents intermediate- 
calcium habitat because it contained both native and agricultural fruits. Village 
agricultural land was considered calcium-poor habitat because it contained only 
agricultural fruits that were nutrient-poor and sparsely distributed in the landscape. 



70 



Table 6-1 Nutrient classification of habitat types in American Samoa. 



Habitat type 


Sample of trees used by 


Calcium 


Habitat classification 




fruit bats* 


content 
(mg/g)** 




Native forest 


Ficus scabra 


10.30 


Calcium-rich 




Planchonella garberi 


4.66 




Mixed agroforest 


Carica papaya 


2.46 


Intermediate calcium 




Myrstica fatua 


n/a 




Village agriculture 


Musa spp. 


0.55 


Calcium-poor 




Artocarpus altilus 


0.91 





* from Whistler 1992, 1994, Cole et al. 1988, Banack 1996. 
** from Nelson et al. 2000a 



Netting of Bats 

Bats were captured using mist nets set at several foraging sites throughout the 
island. Sites were chosen by watching animal movements at dusk, by finding fruit ejecta 
pellet locations, and on the advice of local residents. Bats were captured in large mist nets 
(6-18 m, 4 inch mesh, Avinet, Inc., Dryden NY) attached to pulleys placed high in 
coconut trees (approx. 15 m) or on tall poles. Nets were raised at sunset and were 
checked every 30 minutes until midnight when they were taken down. Netting was 
terminated in the event of rain, a full moon, or excessive wind. Bats were caught from the 
center, east, and west side of the island so that results represent the entire island 
population. After capturing a bat, its sex, reproductive status, overall body condition, 
forearm length, body mass, and time of capture were recorded (Racey 1988). 
Radiotelemetry 

Radiocollars (Model RI-2D, Holohil Systems, Ltd) were fitted around the necks 
of bats using embroidery thread strung inside Tygon rubber tubing (2 mm). The 
transmitters weighed 8.5 g and represented 3% of the bat's body weight. Collars often fell 
off or were removed by bats but were retained for an average of 2.5 months. They were 
recovered, refurbished, and used again. Nocturnal radiotracking involved tracking bats on 



71 



Vainuu Point 




Island Habitat Types 

Mxed Agroforest 
Native Forest 
Village Agriculture 



Fagatele Bay 
5 



K 



10 



15 Kilometers 



Figure 6-1. Map of Tutuila, American Samoa showing the three island habitat types. 
Map follows Freifeld 1998. 

foot or from a truck along roads and trails. The loudest signal method (Springer 1979, 
Kenward 2001) was used to determine the direction to the signal. By sighting down the 
mast of the antennae with a compass, an azimuth was taken. Three azimuths whose 
intersecting angles were generally 30° but no less than 20° apart were used to estimate 
the bat's location. The program Locate (version 2.82, Nams 2001) was used to generate 
the location estimates. 

Bats were tracked two to three nights a week and only once per night to maintain 
independence of locations (Erickson et al. 2001, Kenward 2001). Locations were 
collected in the first six hours of the night (18:00-24:00 h), beginning shortly after bats 
left the roost to forage and continuing until midnight. Collection terminated at midnight 
because bats displayed a sharp decrease in activity beginning at midnight and continuing 
until early morning (Banack 1996). The sequence in which individual bats were located 



72 

was random and opportunistic. Locations taken early in the night were preferred to test if 
calcium was sought first, although previous radio-tracking work indicated that P. 
tonganus often used a single foraging area throughout the night (Banack 1996). 

A digital USGS 7.5" map of Tutuila was proportioned and georeferenced by using 
ESRI's Image Analyst and 18 GPS locations taken across the island. The Freifeld 
vegetation types map (Freifeld 1998) was overlaid onto the USGS map and proportioned 
using Image Analyst. The map image was then converted to an Arc View shape file for 
use in habitat analysis. The map was ground-verified at 175 points to check for accuracy 
of the resulting map. The three areas that received the most bat use were mapped with a 
GPS, converted to GIS, and used for the habitat analysis. Habitat use was determined by 
the number of times each bat was located within each habitat type. The Chi-square 
goodness of fit test and simultaneous Bonferroni confidence intervals (Neu et al. 1974) 
was used to determine if habitat types were used in proportion to their availability (see 
Erickson et al. 2001). Tests of habitat selection were performed for three groups; males, 
reproductive females, and non-reproductive females. To evaluate differences in foraging 
distances flown by bats, SPSS (Norusis 1993) was used. A Kolmogorov-Smirnov test 
revealed that the data were not normally distributed, so the non-parametric Mann- 
Whitney test was used to analyze the data (Sokal and Rohlf 1995). 
Radiotelemetry Error 

Data obtained by two independent observers were evaluated for telemetry error 
using the Location Error method (Zimmerman and Powell 1995) and error ellipse results 
from the Locate program. Fourteen test collars were placed in different habitat types and 
at different distances around the study area to account for the different effects each 
habitat type and distance may have on the radio signal. True collar locations were 



73 

determined using a handheld GPS (GeoExplorer 2, Trimble Navigation, Ltd., Sunnyvale, 
CA). Error was assessed for both observers independently taking radiotelemetry 
locations. A two-sample t-test (Sokal and Rohlf 1995) was performed on the results of 27 
collar locations to evaluate the potential difference in performance between the 
individuals radio-locating bats. The data were transformed by natural log to approximate 
a normal distribution. 

Results 
Radiotelemetry Error 

The results of the t-test showed no difference (p = 0.27) in the error distances 
between the two observers, thus the results were pooled to produce an overall study error 
distance for all radio-location estimates. The mean error for the straight-line distance 
between a known location and a location estimate was 103 ± 138 m. For comparison, the 
mean error ellipse from the Locate location estimate was 1.72 ± 2.63 ha with a median of 
0.54 ha. The standard deviation of 2.63 ha is less than a 100 m straight-line distance. 
Habitat Selection 

Radiocollars were attached to twenty (1 1 males, 9 females) Tongan flying foxes. 
Two of the nine females were lactating and thus classified as reproductive. All other 
females were considered non-reproductive. Seven months of radiotelemetry resulted in 
166 usable locations. Bats left the roost to begin foraging at approximately 18:00. 
Radiotelemetry locations were taken between 18:00 and 24:00 h. (Figure 6-2) 



74 




0000O\CN©Or-irHrir*r*irr>o 

Time of night (h) 

Figure 6-2. Frequency of P. tonganus location estimates recorded from 18:00 to 0:00 on 
Tutuila, American Samoa. 

Mixed agroforest provided 70% of all radiotelemetry locations (1 16/166 

locations) and was preferred (p < 0.001) over the native forest or village agricultural 

habitat types (Table 6-2). Mixed agroforest was used by both sexes and all ages of bats 

across a wide spectrum of time. Native forest was avoided by bats (p = 0.05) and was 

used only 20% of the time. Village agriculture was used only by non-reproductive 

females (n= 9 locations), and never by males or reproductive females. Reproductive 

females (n=2) were located only within mixed agricultural areas. A broken collar on one 

of the females prevented us from getting a greater number of locations for reproductive 

females (n = 10). Eight dropped collars were found; five in mixed agricultural areas, two 

in village agricultural areas, and one in native forest habitat over the seven months. 



75 



Table 6-2. Summary of goodness-of-fit tests for habitat selection for radio-collared 
Tongan flying foxes on Tutuila, American Samoa. 





n 


X 2 


DF 




Habitat type 3 






Village 
agriculture 


Mixed 
agro forest 


Native 
forest 


All bats 


20 


117.18* 


19 


NS 


Preferred 


Avoided 


Males 


11 


69.43* 


10 


Avoided 


Preferred 


Avoided 


Non-reproductive 


7 


41.69* 


6 


Preferred 


NS 


Avoided 


females 














Reproductive 


2 


87.70* 


1 


Avoided 


Preferred 


Avoided 


females 















* significant at p < 0.001, indicating that habitats were not used in proportion to 
availability. 

a Avoid = habitat used less than expected based on its availability 
Prefer = habitat used more than expected based on its availability 
NS= no selection, habitat used in proportion to its availability 



« 100 



s? 




■ Native forest 

■ Mixed agro forest 
□ Village agricultural 



Males Non-reproductive Reproductive 

females females 



Figure 6-3. Percent of locations by habitat type on Tutuila, American Samoa used by 
Tongan flying fox males (n= 1 1), non-reproductive females (n = 7), and reproductive 
females (n=2). 



76 



Distance Traveled from Roost to Foraging Site 

All bats used in this study either roosted on the southwest side of the island at 
Fagatele Bay on Mataautuloa Ridge, or on the northeast side of the island near Afono 
Village at Vainuu Point on Ogetu Ridge (Figure 6-1). Each bat was tracked for an 
average of 2.5 months before the collar fell off or we were unable to locate the bat on the 
island (Table 6-3). The average flight distance for bats from roost to foraging site was 1.8 
km. Males and females did not differ in the distance flown from roost to foraging site (1.5 
km males, 2.3 km females). Both reproductive females were identical in their average 
foraging distance (0.7 km). Although their average distance flown was much less than 
that of other females (0.7 km vs 2.3 km, respectively), the results were not significantly 
different (p = 0.55), most likely due to the small sample size of lactating females (n=2). 

Bats from roosts on the west side of the island (at Fagatele Bay, n= 15) flew an 
average of 0.87 km to their feeding locations, while bats on the east side (at Vainuu 
Point, n=5) flew an average of 4.84 km (Figure 6-4). These foraging distances were 
significantly different (p = 0.002). The single longest straight-line distance flown by a bat 
from the east-side roost (Vainuu Point) was 16km by an adult female. The longest 
straight-line distance flown from a bat from the west-side roost (Fagatele Bay) was 8.1 
km by a young adult female. 



77 





d 




.2 «t 




•*-> ^^ 




03 '2 




o ^ 




© i 




Z H 




o S 




o o 




ftS 




ao 




c 








'5b 




03 <~- 












O CD 




** O 




11 




o w 




B id 

. ^3 




<L) 




> 




< 


ci 




o 


CO 


H 


D 


cd 


>< 


00 


<*H •« 


u 

o 


>> 

i- a 

cd u 


CO 


11 


1 


zl 


(3 




•1-H 


•3 


CO 


03 


03 




XI 




-*-» 




1 


u r3 


u 


CD ^ 


§ 


-° 5 




£ J* 


*o 


3 — 


o 


1 g 


o 

■3 


s 


o3 




c 




c 




03 




ao 




5 


H 


o 


ao 


H 


< 


o 




(N 




CD 




X 




+- ► 




)-H 




a 


X 


| 




i 




1 




3 




</> 


"3 


03 


3 


03 


•o 


Q 


> 






en 
i 

u 


■a 



CO 

6 



CO CO 
1) CD 



CO 
CD 



CO CO 

o D 



eg 
B 



CO 

a 



to 
03 



CO 



m 


CO 


CO 


co 


to 


o 


o 


1) 


(U 


CD 



££££w£££££w£www£££££ 



OS 

© 



© o 



OrvioOOOOt^r^cNc^cNOOOOO 



H 



s <^ *z 



r- 2 *o "*■ 



OOOiA^H^^S^lNSrtf 1 



NNNK^NNNNN^ONCINrttClNH 



3 


3 


3 


■a 


T3 


•5 


■ 


03 


03 


X) 


X 


X 


3 


3 


3 


CO 


oo 


00 



3 3 3 

-i-> +-> *o Td "O 

*3 *3 o3 o3 03 

A A •£ -o x> 

«3 ^3 3 3 3 



3 3 3 3 3 

+j fl fl "O -*_. *j*0"0 
F 7^o3o3o3'T3*7303o3 

^333^^33 



3 3 

"8 -o 



3 3 
T3 73 



<I<!Mty3t/l<;[Zl(/lW<;<t/3C/l<'<-<'< 



fa [JL, 



fa 



§ 2 S S PL, 



til fa 



s s a 



fa 



«-i CN f> 



IT) VO 



00 ON 



O ^ N n <t m vo 



00 ON O 

"-i ^ <N 



79 






I 




m 




B 




O 




1 




■g 




§ 








1 




i 




^K 


1 S 




Ih 


^--* — 7^1 tt S 


£ S 




<2 


/^ > T*n/ 


s g 




<D 


/ \jL> \ 


£ J 

g a 3 




+-* 
M 

O 

o 




a g *. 




E 




I -8 'f 




43 






ro X J 




1 

6 


V ji^^^l— *^^^1 £N> > 1 Jr 


i Q ■ 




e 

o 




o 


\. jtL^BK-'J . ''' .*<L/ m '§*■ 


d 


^** v i % _ ^Bi^i \ ^— »St ^^^»* i ' 


<L> 


4ij I' 1 'i^^E^BriL-i 


1 


/? ** PP 1 ^. 


C/3 


s : ^^x ! 


(L) 

£ 






■f^ 1 


00 

s 
1 

o 

M 

03 




• 

s 1 




k>% 




t «t 9 




^H0 A w^f 


O 




o * =■ 






E 




I 2 | 


M rwsfflP--— ■"'' 


CO 

go 




io » 5 — 




cd o 




9 m S B 
(i je z .» 




•9 © 






p] P7 


1 J3 

E o 




1DM 


\,^i ^t^\ 


J3 E 






'\^P_ ^* O 


J ]J 


^^^v 






o w 




o, ^ 




c?-D 




51 




< | 




1 s 




VO <u 




g i 




3 a 




•2? 2 










(Ih "9 



80 

Discussion 

Radio-tracking studies can highlight what habitat types are preferred by flying 
foxes, and if reproductive females prefer nutrient rich areas for foraging. In this study, 
Tongan flying foxes preferred the mixed agroforest habitat type to all other habitat types. 
Mixed agroforest represented 70% of all radiotelemetry locations. Non-reproductive 
females were the only group that preferred to forage in village agricultural areas, and 
reproductive females foraged only within mixed agricultural areas. Bats roosting on the 
east side of the island flew farther to forage than bats roosting on the west side of the 
island. Reproductive females flew less distance than non-reproductive females to forage. 
Habitat Preference 

The observed preference that P. tonganus displayed for mixed agroforest habitat 
in this study is consistent with other studies on this species (Pierson et al. 1992, Wilson 
and Engbring 1992, Banack 1996, Brooke 1998). The results of this landscape-level 
study are also in agreement with the results of fruit preference tests at the level of the 
individual bat, where agricultural fruits were overwhelmingly preferred to native fruits in 
feeding trials (Chapter 3). Dropped collar locations further support a preference for the 
agroforest habitat type and may be a reflection of time spent foraging in that habitat type. 

Tongan flying foxes exhibited variation in habitat use among males and females. 
Non-reproductive females were the only individuals to use village agricultural areas for 
foraging. Reproductive (lactating) females only fed in mixed agroforest areas. In contrast, 
males were never located in village agricultural areas. In Australia, Spencer and Fleming 
(1989) also found that females {Nyctimene robinsoni) were more likely to feed on fruit in 
fruit orchards than males, whereas males preferred to feed on native figs. Because figs are 
calcium rich (O'Brien et al. 1998, Nelson et al. 2000a), and reproductive or parous 



81 

females may be calcium deficient, it seems counterintuitive for males, but not females, to 
prefer them. Elangovan et al (2001) found that carbohydrates and water were consumed 
first following the extended day roosting period. Females may be more water or energy 
stressed than males, particularly females while lactating. Agricultural fruits tend to be 
both juicier and more sugar-rich than native fruits (Oftedal and Allen 1996). This may 
contribute to a preference for agricultural fruits, particularly among reproductive females 
that emerge following a period of fasting in their day roost and demands associated with 
lactation. 
Distance Flown from Roost to Foraging Site 

Flying fox species typically fly 10-30 km between roost sites and feeding 
locations (Mickleburgh et al. 1992, Palmer and Woinarski 1999). Previous research in 
American Samoa indicates that Tongan flying foxes are capable of flying 45 km in a 
single night (Banack 1996). While a single adult female in this study flew 16 km from a 
roost to a foraging location, this behavior was not typical of the group. Instead, Tongan 
flying foxes in this study averaged less than 2 km flying distance between their roost and 
foraging locations. Bats flew approximately 1/10 the distance reported for P. tonganus in 
1992-1994 on the same island (Banack 1996). 

Differences in the distances flown by P. tonganus may be related to the years in 
which the studies were done. Banack's study was conducted two years after a series of 
destructive hurricanes battered the Samoan archipelago. With sustained winds in excess 
of 200 km/hr, the hurricanes stripped trees of their fruit and leaves, and severely reduced 
the food base of the island (Elmqvist et al. 1994, Pierson et al. 1996). For several years 
following the hurricanes, food was scarce on Tutuila, and Tongan flying foxes had to fly 



82 

farther to find food (see Craig et al. 1994b, Nelson et al. 2000b). In the decade since the 
hurricanes, food has become plentiful on the island. Monthly fruit surveys conducted in 
2001 demonstrated that agricultural fruits were abundant year round on Tutuila (S. 
Nelson, unpublished data, Trail 1994). The year-round abundance and availability of 
agricultural fruits compared to native fruits may have resulted in both a preference for 
them and shorter foraging distances from the roost to find them. Foraging distances have 
been shown to decrease in other bat species when resources were plentiful and increase 
when resources were scarce (Spencer and Fleming 1989, Palmer and Woinarski 1999). 
Foraging Distance and Roost Affiliation 

The greatest differences in foraging distances among P. tonganus were related to 
roost affiliation. Bats that roosted on the east-side of the island traveled an average of 5.5 
km farther to forage than did the bats that roosted on the west side. Both roosts were in 
native habitat, as is typical of this species (Brooke 1998). However, the roost on the west 
side of the island (at Fagatele Bay) was near a large agricultural area. In contrast, bats on 
the east side of the island were near the National Park of American Samoa, which is 
primarily native forest habitat (Cole et al. 1988, Whistler 1994). Although native forest 
provided a nutritionally richer resource, bats from the east side flew across the island 
nightly to feed in mixed agroforest areas. For example, an adult female flew an average 
of 9 km each night from her east-side roost to feed within different mixed agroforest 
areas on the west side. Other flying foxes that roost in native forests but feed in 
agricultural areas also fly long distances to reach the agricultural areas (Mildenstein 
2002). 



83 

Foraging Patterns of Reproductive Female Bats 

Reproductive females exhibited atypical foraging behaviors when compared to 
non-reproductive female bats. The two lactating females traveled less than any other 
group of bats. Both lactating females roosted on the east side of the island (Fagatele Bay) 
but flew less than the average foraging distance for other bats that also occupied that 
roost. These two females also flew identical distances (0.7 km) and flew the same 
distance each night they were tracked to foraging areas (n=10). Foraging distances for 
reproductive females may be restricted to areas near the maternity roost because they 
have dependent young (Palmer and Woinarski 1999). This may explain the consistency in 
their flight distances and the relatively short distances flown by both reproductive 
females. Dominque (1991) found that pregnant and lactating Carollia perspicillata 
females flew almost as many flying bouts as non-reproductive females, but their flights 
were much shorter. In contrast, non-reproductive females performed longer exploratory 
flights to survey for fruit abundance. Thus, the feeding behavior seen in reproductive 
females may be a means of shifting energy from exploratory behavior to reproductive 
effort (Dominique 1991). 
Nutritional Landscape Ecology 

Little is known about the kind of information available to animals at the scale of 
ecological landscapes, and how this information is used with respect to habitat selection 
(Lima and Zollner 1996). The central prediction of this study was that reproductive 
females would forage in calcium-rich habitats to relieve calcium deficiencies that 
potentially arise during pregnancy and lactation (Keeler and Studier 1992, Radostits et al. 
1994, Studier et al. 1994a, Bernard and Davison 1996). However, a strong preference for 
calcium-poor agricultural habitat areas and an avoidance of calcium-rich native forest 



84 

habitats suggests that Tongan flying foxes may forage to maximize the intake of nutrients 
other than calcium. Agricultural fruits were selected despite the calcium deficiency that 
resulted from consuming them, based on an assumption of the standard (Chapter 3,5). 
While foraging, Tongan flying foxes appeared to seek out fruits that were high in energy- 
rich carbohydrates (sugar), similar to what was reported for Cynopterus sphinx 
(Elangovan et al. 2001, Chapter 3, 7). 

A potential limitation of this study was the bias toward documenting initial 
foraging flights. Elangovan et al. (2001) found that C. sphinx fed on predominantly 
energy-rich fruits during the early hours of the night, and foraged for concentrated 
mineral sources later in the night. Sugar-rich agricultural fruits provide the highest energy 
return for an animal's foraging effort, and may relieve a carbohydrate and water debt 
incurred while at the day roost (Kurta et al. 1989, Elangovan et al. 2001). Initial foraging 
flights were prioritized under the assumption that bats would first forage to relieve 
calcium deficiencies. Early night was also chosen to avoid the period of rest typical off. 
tonganus later in the night (Banack 1996). In addition, because all lands are private in 
American Samoa, work was terminated at midnight to avoid disturbing residents sleeping 
in open houses (fales) while tracking bats on private land. However, Banack (1996) 
found that most P. tonganus foraged within a single area throughout the night, following 
a period of foraging upon arrival in the area (Banack 1996). In another study, C. sphinx 
left the roost to begin foraging at 18:00, and commenced leaf-eating at 19:30 h 
(Elangovan et al. 2001). Almost 90% of the radiotelemetry locations in this study were 
recorded after 19:30. Thus, these radiotelemetry locations may include leaf-eating, and 
reflect the consumption of high-energy fruits earlier and leaves later in the night. 



85 

Due to the paucity of alternative vertebrate pollinators and seed dispersers, fruit 
bats are considered keystone species on Tutuila (Cox et al. 1991, Banack 1998, Rainey et 
al. 1995). A decline in bat population size could affect community structure and 
biodiversity on the island. Mixed agroforests were highly preferred by Tongan fruit bats, 
and their maintenance is important for successful foraging. P. tongnanus and its congener 
on Tutuila, Pteropus samoensis, are both dependent on native forest for roosting (Brooke 
1998, 2001). It is critical to preserve native forest on Tutuila to maintain roosting habitat 
and to create a buffer from anthropomorphic disturbance. Disturbance of maternal roosts 
can lead to abandonment of the roost and have population-level effects (Brooke 1998). 

In summary, Tongan flying foxes showed a strong preference for agricultural 
habitats. Bats commuted nightly over native forest habitat to feed in mixed agroforest 
areas or fed in adjacent mixed agroforest areas near their roosts. Tongan flying foxes 
appeared to forage in a manner that maximized their energy intake rather than their 
calcium intake. Succulent and sugar-rich agricultural fruits were possibly preferred by 
hungry and dehydrated bats emerging from the day roost as a source of quick, high- 
energy food. Agricultural fruits were plentiful and available year-round on Tutuila, which 
may have resulted in minimum foraging distances to find them. Reproductive females 
may be constrained to forage near the maternity roost to support energy-demanding 
lactation and, if so, reproductive females should prefer high-energy, locally abundant 
agricultural fruits. Future habitat use studies of flying foxes should include radiotelemetry 
locations taken throughout the night and an investigation of foraging within a larger 
group of reproductive females. Together, these studies may prove decisive in determining 
if P. tonganus attempts to increase their calcium intake by feeding in nutrient-rich areas. 



CHAPTER 7 
ABSORPTION AND UTILIZATION OF MPNERALS CONSUMED BY CAPTIVE 
LACTATING FEMALE MALAYAN FLYING FOXES (PTEROPUS VAMPYRUS) 

AND THEIR PUPS 

Introduction 

Fruit bats of the suborder Megachiroptera and family Pteropodidae are 
increasingly bred and maintained in captivity. However, little is known about either their 
mineral requirements or the adequacy of the diets fed to them. Because nutritional 
standards are unknown, dietary recommendations for fruit bats have been based on 
standards for rats, averages for all mammals, or diets that maintain breeding colonies of 
flying foxes in captivity (Fascione 1995, NRC 1995, Dierenfield and Seyjagat 2000b). 

Two separate factors, the overfeeding of heterogeneous diets, and dominance 
hierarchies in social species, are now recognized as important factors influencing the 
nutritional intake of captive wildlife (Robbins 1993). The distribution and composition of 
the daily diet given to fruit bats in captivity are quite different than that consumed by 
free-ranging bats. In the wild, Pteropus vampyrus flys up to 50 km each night to reach its 
feeding grounds, and the temporal and spatial distribution of food resources are complex, 
so that food acquisition often requires a large proportion of an animal's time budget 
(Medway 1969, Oftedal and Allen 1996, Kunz and Jones 2000). In addition, wild animals 
do not eat more than they require because this effects their wing loading and ability to fly 
(Dudley and Vermeij 1992). In captivity, energy requirements are less because bats do 
not travel to their food source, and flight is restricted by cage size (Courts and Feistner 
2000). Thus, the combination of reduced activity and plentiful food result in captive bats 

86 



87 

that are heavier than their wild counterparts. This may lead to obesity in dominant 
individuals (Allen and Oftedal 1996, LeBlanc 1999). 

When captive animals are housed in groups, it is common practice to feed 
amounts somewhat in excess of consumption to ensure that all individuals have access to 
food, and that young and subordinate animals receive adequate quantities (Courts and 
Feistner 2000). Animals often choose among the numerous food items offered, and may 
ingest a diet that is much different from the diet that was offered. Therefore, assessments 
of the nutritional adequacy of a diet should be based on what is actually eaten rather than 
what is offered (Oftedal and Allen 1996). Intake levels also vary with origin, age, 
dominance, and reproductive status of animals (Courts and Feistner 2000). 
To accurately document mineral intake, this study evaluated mineral absorption 
efficiencies using apparent absorption. This method measured both mineral intake and 
excretion, and accounted for the unique manner of fruit bat feeding where bats chew the 
plant matter into a fibrous pellet, swallow the juice, and eject the flattened pellet (Lowry 
1989, Kunz and Ingalls 1994, Kunz and Diaz 1995). Previous studies failed to account 
for minerals found in ejected pellets. 

Growth, pregnancy, lactation, age, gender, nutrient interactions and illness can 
influence nutrient requirements. Appropriate amounts of calcium and phosphorus are 
especially critical for bats during early growth and peak lactation (Barclay 1995, Hood et 
al. 2001). If dietary intake of calcium is inadequate, pregnant and/or lactating females 
donate their own skeletal calcium to build the skeletons of their young (Bernard and 
Davison 1996). Inadequate dietary calcium can result in weakened bones and low milk 
production in females (Radostits et al. 1994). Patterns of post-natal growth in the pups are 



88 

influenced by the quantity of milk from the female fruit bat (Kunz and Stern 1995, Kunz 
and Hood 2000). Inadequate calcium results in inhibited growth and reduced 
mineralization of bone in offspring (McDowell 1992, Radostits et al. 1994). Additionally, 
nutritional conditions in captivity may push post-natal growth rates to their maximum, 
adding further nutritional stress (Kunz and Stern 1995, Kunz and Hood 2000). Thus, it is 
critical that adequate minerals are available in the diet of lactating females and their 
rapidly growing offspring. 

In this study, two pens of lactating female P. vampyrus and their rapidly growing 
pups were compared to a control of non-reproductive P. vampyrus females to evaluate 
differences in dietary intake and assimilation of minerals as determined by apparent 
absorption. The large size of the Malayan fruit bats, coupled with the nutrient demands of 
lactation and rapid growth (Barclay 1994, Kunz and Hood 2000), served to illustrate if 
their diet is nutritionally adequate for a breeding population of large captive flying foxes. 

Methods 

This study used Pteropus vampyrus, also called the Malayan flying fox, to 
examine mineral nutrition in captive fruit bats. P. vampyrus is one of the world's largest 
flying foxes, weighing approximately 1 kg and attaining a wingspan of over 2 m (Ingle 
and Heaney 1992, Kunz and Jones 2000). Males generally weigh more than females, and 
overweight males in captivity can weigh as much as 1.5 kg. Populations of female P. 
vampyrus give birth synchronously during a single annual peak, usually to a single 
offspring (Mickelburgh et al. 1992). The gestation period is approximately 180 days after 
which young suckle from their mothers for 2-3 months, but young may depend on their 
mothers for as long as a year (Pierson and Rainey 1992, Kunz and Jones 2000). 



89 

Three pens of P. vampyrus were tested within their normal outdoor flight cages at 
the Lubee Foundation, Inc. in Gainesville, Florida in January 2000. Two pens (pens 1 and 
2) consisted of lactating female P. vampyrus and their rapidly growing pups. A separate 
pen (pen 3) served as a control and contained only non-breeding females. Bats were kept 
in octagonal outdoor flight cages that contained an enclosed indoor portion (the core) 
where all food was suspended and consumed. Bats were fed their normal diets (Table 7- 
1) and were fully acclimated to both their surroundings and diet for two months before 
the experiment began. Bats were fed approximately one-half their body mass per day 
(wet weight). Stainless steel bowls of food were hung near the top of the core to prevent 
fecal and urinary contamination. Food was available to the bats from 1 500 to 0900 h each 
day. Water and salt licks were available to the bats at all times. Salt licks were composed 
of salt and mineral oil and contained 96-99 % salt (Pet Products, Inc., Hauppauge, N.Y.). 
Water samples contained 0.10 mg calcium, 0.20 mg phosphorus, 0.3 mg magnesium, 
0.05 mg potassium, 0.03 mg zinc, 0.20 mg iron, 0.30 mg manganese, and 0.31 mg 
sodium for every 1 ml of water offered to the bats. Copper levels were not detectable. 



Table 7-1. Composition of food fed to P. vampyrus at the Lubee Foundation, Inc. 

Fruit type Percent of diet (%) 

Apple 35.9 

Pear 6.8 

Banana 13.6 

Juicy fruit (grapes, kiwi, pineapple) 7.8 

Mixed fruit (seasonal items) 2.6 

Cantaloupe 12.0 

Carrot/sweet potato (steamed or raw) 9. 1 

Kale/spinach 6.5 

Dry Zupreem primate biscuit* 5.0 

Vitamin E solution 0.1 

Dicalcium phosphate 0.6 



"this was replaced with Lubee Fruit Bat Supplement in February 2001. 



90 

Two days of trials were performed in the three pens. Upon arriving at the indoor 
enclosures the next morning, rejected food and feces littered the enclosure floor. Food 
remains were separated into three categories: 1 . uneaten food that remained in the bowls 
and rejected food on the floor (orts), 2. food that had been masticated, the juice 
swallowed, and the fiber pellet rejected (ejecta), and 3. fecal remains (feces). Samples of 
orts, ejecta, and feces were collected by scraping the samples off the floor and walls 
using a sterile plastic paint scraper or picking them up using stainless steel tweezers. Both 
the scraper and tweezers were rinsed in deionized water between samples. Latex gloves 
were worn during collection to further prevent contamination. Some of the orts and ejecta 
samples on the floor of the enclosure were contaminated with feces. Those samples were 
sprayed with distilled water to remove as much fecal matter as possible. Fresh masses 
were recorded and subsamples were pooled for each pen and frozen for future analysis. 
Dehydration factors for all subsamples were subtracted from wet weight values. 

Samples were analyzed at the Animal Nutrition Laboratory at the University of 
Florida in Gainesville, Florida. Frozen samples were thawed and placed in a drying oven 
for 24 h at 105° C. Dried samples were weighed and ashed at 550°C for 12 h. Each 
sample was prepared and digested according to the procedures of Miles et al. (2001). 
Following wet digestion, calcium (Ca), magnesium (Mg), potassium (K), sodium (Na), 
iron (Fe), copper (Cu), zinc (Zn), and manganese (Mn) contents were assessed by flame 
atomic absorption spectrophotometry using a Perkin-Elmer AAS 5000 (Perkin-Elmer 
1980). Phosphorus was measured separately using a colorimetric assay (Harris and Popat 
1954). All samples were analyzed in duplicate. Standard reference material (citrus leaves 



91 

1572, National Institute of Standards and Technology, Gaithersburg, MD) was analyzed 
with each sample set for accuracy. 

The combination of uneaten food in the bowls and food ejected was subtracted 
from total food offered to obtain a value for total food intake. The mineral concentration 
of fruit was multiplied by the amount of food that was consumed each day to calculate 
total mineral intake. Total mineral intake was used to calculate values for apparent 
absorption. Both equations are presented below. All calculations are on a dry matter 
basis. Dry matter numbers may be slightly lower than the actual fruit value for all fruits 
because the fruit was in the heated core, which may have desiccated the samples. 

Total mineral intake (g) = 

(total amount of fruit offered (g)) - (total amount of orts (g)) - (total amount of 
ejecta (g)) 

Apparent absorption (%) = 

(total mineral intake) - (fecal mineral concentration) 
(total mineral intake) x 100 

All values for diet, ejecta, orts, fecal, intake, retention, and apparent absorption 
were statistically analyzed using the General Linear Model (GLM) procedure in SAS 
(SAS 1999). This procedure was used to test for a difference in the means among the 
three pens. Significance was evaluated at the p < 0.01 level because of the large number 
of comparisons tested, and that the values were not all independent measurements, but 
were based on previous values (e.g., retention and absorption) (R. Littell, pers.comm.). 

Results 

Nineteen lactating females and sixteen pups were compared to twelve non- 
breeding female P. vampyrus to evaluate dietary differences between the groups (Table 
7-2). Three of the females in the lactating pens were without pups due to the death of two 
pups and one infertile female. The pups were from a single breeding cohort and were all 



92 

6-8 months old. P. vampyrus pups at the Lubee Foundation, Inc. are usually in the 
weaning process at four to six months of age, but they will continue to nurse 
opportunistically until they are eight months old (D. LeBlanc, pers. comm.). Many fruit 
bat species continue to nurse at one year of age (Galindo et al. 1995, West 1986). 

To compensate for the demands of prolonged lactation in females and growth in 
pups (Radostits et al. 1994), females in pens 1 and 2 were given twice the amount of food 
(wet mass) that the non-breeding females received in pen 3. The two groups of lactating 
females and their offspring were very similar in the dry diet consumed as a percent of 
body mass (1 1.2% and 10.5% respectively). Non-breeding females consumed 
approximately 3% less than the lactating females with pups. Lactating females 
presumably ingested more food to compensate for the energy demands of lactation. 



Table 7-2. A comparison of lactating female P. vampyrus with pups to nonbreeding 
females for number of bats in pen, diet offered (wet mass), and dry matter intake as a 
percentage of body mass. M: male, F: female, N/A, data not available, pups were nursing. 



Pen 


Treatment group 


Number of bats 
in pen 


Diet offered 


Dry matter intake as 
% of body mass 


Penl 


Lactating females 


10 


H,314g 


11.2% 




Number of pups 


4M,5F 


N/A 


N/A 


Pen 2 


Lactating females 


9 


10,565g 


10.5% 




Number of pups 


4M, 3F 


N/A 


N/A 


Pen 3 


Nonbreeding females 


12 


5357g* 


7.2% 



indicates a significant difference between pen 3 when compared to pens 1 and 2. 
The mineral composition of the diet offered and consumed by the bats was 
compared to the minimal mineral requirements reported for other mammals (Table 7-3). 
The minimum values for calcium and phosphorus followed the recommendation for fruit 
bats by the American Zoo and Aquarium Association Chiropteran Taxon Advisory Group 
(AZA Bat Tag) (Fascione 1995). All remaining mineral values used are based on 



93 

requirements for rats and were consistent with mammalian standards (Oftedal and Allen 
1996). Both the diet offered and diet consumed in this study consistently equaled or 
exceeded the recommended requirements for all minerals examined. There was a 
difference among the minerals offered and the minerals consumed for the majority of 
minerals studied (Table 7-3). The amount of minerals offered is higher than that 
consumed, suggesting that bats consumed the lower nutrient foods from the diet and 
rejected the high nutrient foods. 

Table 7-3. A comparison of the diet offered and consumed by P. vampyrus in this study 
to the recommendations of the American Zoo and Aquarium Association Chiropteran 
Tax on Advisory Group Bat and the mammalian standard.* 

Ca P Mg K Na Fe Cu Zn Mn Ca : P 

Mineral % % % % % ppm ppm ppm ppm ratio 

requirement* 0,5 0.4 0.06 0.4 0.06 39 6 11 11 1:1-1:2 

Diet Offered 0.9 0.7 0.1 0.9 0.2 93 8 ~^9 25 1.3:1 
Diet Consumed 

Penl 0.6 0.5 0.1 0.6 

Pen 2 0.5 0.4 0.1 0.5 

Control 0.6 0.5 0.1 0.5 

Requirements for Ca and P are from Fascione 1995, all others are rat values from Allen 

and Oftedal 1996. 

All values are averages for two days of collection in each pen, ppm = parts per million. 

Because imbalances in the ratio of Ca to P leads to poor absorption and 
exacerbate deficiencies of both minerals, preferred dietary levels are typically expressed 
at a desired range of 1 :1 - 2:1 for the Ca:P ratio (McDowell 1992). The diet offered and 
consumed by bats in this study met or exceeded the 1:1 Ca:P ratio. 

The mineral composition of the diet offered to bats in pens 1 and 2 was identical 
(Table 7-4). The non-breeding females in Pen 3 received significantly less Ca, P, Mg, K, 
and Na than the lactating groups, presumably because the volume of diet offered in pen 3 



0.1 


64 


5 


34 


19 


1.2:1 


0.1 


57 


5 


30 


18 


1.2:1 


0.1 


70 


6 


36 


19 


1.2:1 



94 

was half of the volume offered in pens 1 and 2 and significantly lower (p < 0.01) for all 
macrominerals studied (Table 7-2). Rejected food, or orts, differed (p< 0.01) in mineral 
concentration for every mineral but Mn between the two pens containing lactating 
females. Given that they were offered an identical diet, this suggests that the females of 
pen 2 rejected more of the nutrient rich foods such as sweet potatoes, carrots and kale 
than the females in pen 1. The control group was significantly lower (p < 0.01) than pens 

1 and 2 for all minerals studied. This reflects the fact that Pen 3 did not have any leftover 
food. This is a term called licked-bowl-clean in captivity (D. LeBlanc, pers.comm.). If it 
persists over several days, the diet volume offered is often increased. However, the 
control bats were overweight and did not need additional food, but overate as is often 
typical of non-breeding captive bats (Courts and Feistner 2000). 

The mineral content of the feces of all three pens was homogeneous. Some 
minerals such as Ca are highly conserved in animals and appear only in minimal amounts 
in the feces when they are deficient in the diet (Ammerman 1995). Despite the ingestion 
of only half the minerals by the control bats in pen 3, their fecal matter contained similar 
amounts of minerals to the bats in pens 1 and 2. This suggests that the bats in pen 3 were 
not in mineral stress. Similarly, mineral retention was very similar for bats in pens 1 and 

2 but were significantly different (p < 0.1) for those housed in pen 3. Mineral intake was 
presumably lower for bats in pen 3 because the diet volume placed in that pen was half 
that given to the other pens. Lactating females were significantly higher (p < 0.1) than the 
control females for mineral apparent absorption of most of the minerals examined in this 
study, excluding Cu and Na. 



95 



Table 7-4. A comparison of the mineral amounts for diet, oris, ejecta, mineral intake, 
mineral retention, and apparent absorption of minerals between two pens (1,2) of 
reproductive females P. vampyrus and one pen (3) of non-reproductive female 
P.vampyrus. 





Pen 


Ca 


P 


Mg 


Na 


K 


Cu 


Mn 


Fe 


Zn 




mg/g 


mg/g 


mg/g 


mg/g 


mg/g 


mg/g 


mg/g 


mg/g 


mg/g 


mg/g 


Diet 


1 


1151 


901 


153 


250 


1107 


1 


3 


12 


6 


Offered 


2 


1276 


1000 


170 


277 


1227 


1 


3 


13 


7 




3 


863* 


676* 


115* 


187* 


830* 


1 


2 


9 


5 


Orts 


1 


137 § 


99 


17 f 


29 § 


111 s 


0.1 * 


0.3 


1" 


i« 




2 


313 


223 


36 


66 


230 


0.2 


0.3 


3 


2 




3 


0* 


0* 


0* 


0* 


0* 


0* 


0* 


0* 


0* 


Ejecta 


1 


229* 


180 


37 


63 


284 


0.3 


0.5 


2 


1 




2 


194 


174 


33 


56 


285 


0.2 


0.3 


2 


1 




3 


237* 


370* 


40 


62 


304 


0.2 


0.6 


2 


1 


Feces 


1 


224 


124 


28 


27 


153 


0.2 


0.6 


2 


2 




2 


250 


147 


27 


22 


119 


0.2 


0.8 


3 


3 




3 


280 


105 


30 


17 


144 


0.2 


0.7 


3 


2 


Mineral 


1 


785 


622 


99 


158 


711 


0.7 


2 


8 


4 


Intake 


2 


769 


602 


100 


155 


713 


0.7 


3 


8 


4 




3 


625* 


490* 


75 


125* 


526* 


0.6 


2 


7 


3 


Mineral 


1 


560 


468 


72 


131 


559 


0.5 


2 


6 


2 


Retention 


2 


518 


438 


73 


133 


594 


0.5 


2 


5 


2 




3 


345* 


372* 


45* 


108* 


382* 


0.4 


1* 


4 


1* 


Apparent 


1 


71 


72 


72 


83 


79 


72 


74 


69 


44 * 


Absorption 


2 


68 


66 


73 


85 


83 


74 


71 


63 


38 


% 


3 


56* 


49* 


60* 


87 


72* 


71 


62* 


55* 


33* 



§ indicates a significant (p<0.01) difference between pens 1 and 2 
* indicates a significant (p<0.01) difference between pen 3 when compared to pens 1 and 
2. Pens 1 and 2 contained lactating P. vampyrus females with pups, pen 3 contained non- 
breeding females and served as a control. Mineral values are averaged for both days of 
the trials and all values are reported in mg unless otherwise indicated. 

Discussion 

Both the diet offered and consumed by bats in this captive study was adequate to 
maintain other mammalian species in a non-breeding reproductive state and met the 
recommendations of the AZA Bat Tag for Ca and P (Fascione 1995, Allen and Oftedal 
1996). The discrepancy between the offered nutrient levels and nutrients actually 
consumed suggests that bats chose less nutrient dense foods and rejected mineral rich 



96 

foods. Elevated apparent absorption values indicate that lactating females and their pups 
were mineral stressed for most of the minerals evaluated while consuming a standard diet 
that met nutrient recommendations for non-breeding bats. 

Although the food fed to bats in this study met the recommendations for captive 
fruit bat diets, the mineral requirements for lactating females and rapidly growing young 
fruit bats remain unknown. Requirements for these groups are usually greater than for 
non-reproductive adults and are reflected in higher apparent absorption values (Kung et 
al. 1998). Apparent absorption is a measure of mineral retention expressed as a 
percentage of mineral intake. Higher absorption may indicate a greater physiological 
need (Keen et al. 1986, NRC 1989, Koo and Tsang 1991, Brink et al. 1992, Bronner et al. 
1992). Animals that are deficient in a mineral (e.g., Ca) compensate by absorbing a 
maximum amount from the diet, which results in an elevated mineral absorption value 
(Ammerman 1995). For example, growing children may absorb up to 75% of ingested Ca 
as compared to 20-40% typically absorbed in adults (RDA 1989). In the present study, 
the pens that contained lactating females and their growing offspring had higher apparent 
absorption values for almost all minerals studied when compared to the control group of 
non-reproductive females. Elevated absorption levels indicate that lactating females and 
pups may be mineral stressed compared to the control group of non-breeding females. 
Minerals are most likely deficient in their diet because of the nutrient demands of 
prolonged lactation and rapid growth (Barclay 1995). 

Calcium is of particular importance for lactating females and their rapidly 
growing pups (Kunz and Hood 2000). Young bats are not able to fly or forage 
independently until they have achieved nearly adult dimensions (Barclay 1995). 



97 

Consequently, near the end of lactation, female bats must feed young that are nearly adult 
size (Kunz and Stern 1995, Kunz and Hood 2000). Thus, it is important to provide 
adequate Ca in diets of lactating females. The discrepancy in Ca between the diet offered 
and consumed by bats suggests that P. vampyrus rejects many of the high Ca foods 
offered to them. Foods such as raw sweet potato, kale, and spinach are high in nutrients, 
but are fibrous and low in sugar, and are not normally encountered by wild bats (Oftedal 
and Allen 1996). These foods were consistently rejected by P. vampyrus. These results 
are consistent with the preferences of fruit bats from other studies, where bats preferred 
soft, non-acidic, low nutrient fruits with high sugar content (Parry- Jones and Augee 1991, 
Chapter 3, Courts and Feistner 2000). Therefore, high-Ca foods and Ca supplements 
should be in a form that is actually consumed by P. vampyrus or the Ca will be wasted. 

The specific mix of produce and items fed to captive animals reflects ideas about 
food selection in the wild, food preferences demonstrated in captivity, established feeding 
practices, and local market availability of the food (Oftedal and Allen 1996). The diets 
prepared for fruit bats and other frugivores in captivity usually consist of chopped fruits 
and vegetables, manufactured primate pellets, and supplements such as vitamin E or Ca 
in an attempt to provide a balanced diet (Courts and Feistner 2000). However, captive 
diets also include commercially grown fruits and vegetables that are preferred by humans 
and are high in sugar (Oftedal and Allen 1996). Fruits such as apples and pears are used 
the most frequently and in the greatest quantities in captive diets because they are readily 
commercially available (Courts and Feistner 2000). Apples and pears together were 
almost half (43%) of the food that was fed to bats in this study. Commercial agricultural 



98 

fruits are generally depauperate in minerals and higher in sugar when compared to fruits 
available to flying foxes in the wild (Nelson et al. 2000a). 

Based on the results of this study, the following changes in diet and procedures 
concerning captive pregnant or lactating female bats and their rapidly growing pups for 
P. vampyrus and other flying foxes are proposed. 

1. Reduce the amount of apple in the diet and replace it with nutrient-rich, preferred 
fruits such as figs and papaya . Apples can comprise 36% of captive diets, but are 
nutrient poor and not preferred by bats (Oftedal and Allen 1996, Courts and 
Feistner 2000). Figs provide a very concentrated source of minerals and are a 
dietary staple of P. vampyrus in the wild (O'Brien et al. 2000, Nelson et al. 2000a, 
Stier and Mildenstein 2001). Papaya is a preferred fruit of wild flying foxes in 
American Samoa and it is rich in Ca (Chapter 3). 

2. Increase mineral availability, particularly for Ca, in pens containing pregnant 
and/or lactating females and growing pups . Commercial mineral and Ca licks, Ca 
supplements, preferred tree leaves (Banack 1996, Chapter 4), and high Ca foods 
such as figs will increase mineral availability to supplement the standard diet given 
to all bats (O'Brien et al. 2000, Nelson et al. 2000b). 

3. Do not breed P. vampyrus females every year . Because the Ca demands of raising 
several young in sequential years may result in osteoporosis (Studier et al. 1994a, 
Keeler and Studier 1992), females should have alternate years between births to 
rebuild their mineral reserves and the bone matrix donated to build the skeletons of 
their offspring (Bernard and Davison 1996). 

4. Wean the offspring earlier from their mothers . This would reduce the occurrence 
of opportunistic and almost fully-grown offspring from continuing to suckle from 
their mothers, and would decrease both the volume of milk produced by the female 
and the duration of lactation (McDowell 1992). A shortening of the weaning period 
may reduce the amount of skeletal Ca donated by the female and allow her a longer 
recovery time to rebuild her Ca reserves following lactation (Bernard and Davison 
1996, Keeler and Studier 1992). 

Wildlife nutrition is still a very young science (Robbins 1993). Much research 

remains to be done, particularly on exotic animals such as fruit bats. Easy access to 

plentiful amounts of food in the captivity allow for selection of nutrient-poor, sugar-laden 

foods preferred by bats. Replacement of these fruits with preferred fruits that are also 






99 

high in mineral nutrients may reduce overeating and food waste. Lactating female 
Malayan flying foxes and their pups can survive and reproduce on the diet currently 
recommended and fed in captivity, but they are likely mineral stressed when fed this diet. 
Additional minerals are needed for pregnant, lactating, and rapidly growing animals. A 
greater volume of the standard diet given to each bat in a pen is not sufficient to meet 
their mineral needs. Instead, they require access to mineral rich sources such as figs, 
preferred tree leaves, commercial mineral licks and dietary supplements in addition to 
their standard diet (Chapter 3,4). Access to these rich mineral sources may reduce 
absorption rates and mineral stress in pregnant or lactating bats. If mineral stress in 
captivity can be resolved for lactating female P. vampyrus, one of the world's largest 
flying foxes and therefore the most likely to have the greatest mineral requirements, it is 
assumed that other captive bat species fed a similar diet while pregnant or lactating will 
not experience mineral stress. Future work should evaluate the effectiveness of various 
supplements and dietary changes and their associated effect on mineral absorption levels. 



CHAPTER 8 
SUGAR CONCENTRATION PREFERENCES OF TWO SPECIES OF BLOSSOM- 
BATS (SYCONYCTERIS AUSTRALIS AND MACROGLOSSUS MINIMUS) IN PAPUA 

NEW GUINEA 

Introduction 

Blossom bats are the smallest members of the family Pteropodidae of the Old 
World fruit bats. Both Syconycteris australis, the northern blossom bat, and 
Macroglossus minimus, the southern blossom bat, have large geographic distributions 
including Indonesia, Southeast Asia, Papua New Guinea and northeastern Australia 
(Figure 8-1). Blossom bats have remarkable plasticity in their temperature regulation and 
energetics. They inhabit areas with temperatures ranging from tropical to temperate in 
climate. Blossom bats live in a variety of habitats, including disturbed successional 
forests, primary lowland forest, and montane rain forest (Bonaccorso and McNab 1997). 
The distribution of S. australis extends farther south than any other small Australian 
megachiropteran and borders on areas with a temperate climate (Law 1994). M. minimus 
has a wide distribution but is limited to more tropical areas (Geiser et al. 1996). 

Morphologically, the two species of bats are almost identical in size and mass 
(a. 12-20 g). However, S. australis is distinguished from M. minimus by a more robust 
dentition that is better suited for feeding on fruits and figs (Ficus sp.), in addition to 
nectar and pollen of wild and domestic bananas (Flannery 1995a, Bonaccorso 1998, 
Winklemann et al. 2000). Although a dietary generalist in Papua New Guinea, S. 
australis is a nectar specialist in Australia, and feeds on blossoms of 1 8 rainforest species 
and from flowers of cultivated bananas (Law 1992, 1994, 2001). M. minimus is a nectar 

100 



101 



specialist in Papua New Guinea, preferring the nectar of domesticated bananas, coconuts 
and mangroves (Gould 1978, Heideman and Heaney 1989, Bonaccorso 1998). 

Specimens of M. minimus taken from New Guinea lacked fruit in the stomach 
(McKean 1983) and captive individuals refused to eat fruit when it was offered to them 
(Bonaccorso 1998). 




Figure 8-1. Distribution of blossom bats in Meganesia. Solid line is Macroglossus 
minimus, broken line is Syconycteris australis. From Bonaccorso and McNab 1997. 

Blossom bats are highly suitable for studying preferences in nectar concentration. 
They are small, volant homeotherms with a high metabolic rate and no means of storing 
large energy reserves (Lemke 1984). Such species might be expected to respond 
markedly to subtle changes in resource availability and to choose high quality food areas 
(Law 1995). Several studies have suggested that blossom bat abundance may be limited 
by food availability and that bats are energy-sensitive foragers (Law 1992, 1993b, 1994). 



102 

Law (1993b) performed concentration preference tests on S. australis in Australia 
in which bats were given a choice of several concentrations of honey water. He found 
that blossom bats showed a distinct preference for the 40% solution (97% preference), to 
the exclusion of the less concentrated 10% and 20% solutions. When fed ad libitum, 
captive S. australis consumed 174% of its body mass in nectar each night. This study was 
conducted in Australia, where S. australis ' habitat includes cold, wet winters and dry 
summers (Law 1994). Exposure to the low temperatures (only 6.7-7.7 °C) in the winter 
months requires a substantial amount of energy for thermoregulation, and can make food 
a limiting resource for S. australis (Law 1993a, Geiser et al.1996). Law (1994) calculated 
the metabolic costs for bats in the winter exceeded 4.5 x of their metabolic rate. Field 
metabolic rates of S. australis in Australia were the highest known for a mammal its size; 
more than seven times higher than their basal metabolic rate (Geiser and Coburn 1999). 
When supplemental food was made available, local populations of S. australis increased 
rapidly (Law 1995). Together, these studies suggest that energy may in fact be limiting to 
blossom bats in Australia. 

Most studies of blossom bat metabolism and energetics have been performed on 
S. australis and M. minimus in Australia. Little is known about the energetic stress of 
these two species in Papua New Guinea. The subtropical and temperate climate and 
distinct seasonality in Australia may subject blossom bats to energy stressors not found in 
the more tropical and benign climate of Papua New Guinea. Therefore, my study 
attempted to examine three ideas. First, what nectar concentrations do blossom bats in 
Papua New Guinea prefer? If they are as energy stressed in Papua New Guinea as they 
are in Australia (Law 1993b), they should prefer the most concentrated nectar 






103 

concentrations available. Second, are there sugar preference differences between S. 
australis, a feeding generalist, and M. minimus, a nectar specialist in Papua New Guinea? 
It is predicted that M. minimus, a nectar specialist, would choose the higher 
concentrations because this species is completely dependent on nectar as a food source. 
Lastly, I attempted to identify if there were differences in energy requirements and 
consumption patterns between sexes or age groups of each species. 

Methods 
Nectar Concentrations 

Sugar concentrations for the nectar preference trials were based on field 
measurements of banana nectar in Kau Wildlife Area on 1 July, 4 July, and 12 July 1999. 
Kau Wildlife Area is adjacent to Baitabag Village in Madang Province, Papua New 
Guinea. Kau Wildlife Area is an 800 ha reserve of lowland forest that contains primary 
old growth, successional, and riparian forest (Winklemann et al. 2000). Flowers of 20 
banana plants were monitored in three all-night sessions that lasted from 1 900-0600 h. 
Blossom bats were often seen or heard during nectar sampling periods. Nectar was 
collected non-destructively with a blunt-edged syringe and samples were placed on a 
hand-held sugar refractometer (Model # 300010, Sper Scientific, Scottsdale, AZ) to 
determine sucrose content on a g solute per 1 OOg-solution basis (Bolten et al. 1979). 
Sugar concentration ranged from 5-24% sucrose equivalents with a mean of 13.24 + 
3.75% (SD) (n= 151) but nectar concentrations varied widely during the night (Figure 8- 
2). The nectar concentration values found here are consistent with those found in other 
bat pollinated flowers (Helverson and Reyer 1984, Kress 1985, Itino et al. 1991). From 
these results, three concentrations were selected for the nectar concentration preference 






104 

experiments, 7%, 15% and 30% nectar, because those values are typical of those 
available to blossom bats in their native habitat. 



C 

o 

£ 

c 

O 

c 
o 
O 



2\) - 
1 






18 
1 f* 














io i 




















14 

1 


















YZ ' 
























1U - 

o 































O 




4 
o _ 




z 
- 


1 1 1 1 1 1 1 T — — "] ! , 



J$ «P # J# <$> ^ *£ A§> J$ J$ J$ J$ 

N?- ^ $>' ^' V ^ V V V ^ b< fe ' 



Time (h) 



Figure 8-2. Banana flower nectar concentrations. Concentration values have been 
averaged for three nights. 

Preference Tests 

Twenty blossom bats were used to determine nectar concentration preferences; 1 5 
S. australis papuanus (7 females, 8 males) and 5 M. minimus nanus (2 females, 3 males) 
(Table 8-1). Bats were captured in mist nets at Kau Wildlife Area and transported to the 
Christensen Research Center in cloth bags. Sex, body mass (g) and forearm length (mm) 
were recorded for each bat. Age was determined as adult or juvenile based on body mass, 
forearm length, and fusion of the metacarpal epiphyses (Anthony 1988). Bats were placed 
individually in 1.5 x 1 x 1 m flight cages for four days. Sugar solutions (henceforth 
"nectar") were made by diluting honey with water in a 100 ml graduated cylinder. 
Concentrations were checked to within 1% by a hand-held refractometer (Atago NI #791, 
Vee Gee Scientific, Inc., Kirkland, WA). 



105 

In preference tests, I compared preference for the 15% to the 7% and 30% sugar 
solutions. Two separate trials were run on each bat. The first two days bats were given a 
choice of either 15% or 30% solution, and the second two days a choice between 15% 
and 7% solution. Thirty milliliters of each nectar concentration in a plastic dish were 
placed in each cage in late afternoon and then collected the following morning. The 
position of the feeders was switched daily to eliminate any preference for the left or right 
side. The amount of sugar solution consumed by the bat was determined by subtracting 
the volume remaining the next morning from the original 30 ml offered. A predetermined 
evaporation constant of 2 ml was subtracted from all consumption volumes. Bats were 
released in the morning if they ate less than 10 ml the previous night. Results from two 
days of each trial were averaged to calculate a mean preference value for each trial for 
each bat. A preference index was calculated as the ratio of the consumption of sugar A 
divided by the total consumption (A+B) (Martinez del Rio 1990, Law 1993b). 

A repeated measures test was used to analyze nectar consumption patterns (SAS 
6.12, SAS 1999), because the four consecutive nights of preference tests were not indepe 

ndent measurements. Consumption levels were adjusted for the body mass of the 
bat as a continuous covariate while I tested for interactions between gender, species, and 
reproductive status. There was no significant difference (p = 0.65) between S. australis 
and M. minimus for consumption volumes; thus they were grouped together for 
subsequent analysis. Because there was a significant interaction between gender and age 
(p = 0.002) that affected consumption, I performed pairwise comparisons using Tukey- 
Kramer's multiple comparisons test to determine which pairwise comparisons were 
different among those factors (Sokal and Rohlf 1 995). 



106 



Table 8-1. Bats used in the nectar preference tests. Daily average and total consumption 
of all concentrations combined are based on four nights of experiments. 









Body 


Length of 


Average 


Total 


Species 


Gender 


Group 


mass 


forearm 


consumption 


consumption 








(g) 


(mm) 


(mL) 


(mL) 


S.australis 


Female 


Adult 


19 


44 


23 


92 


S.australis 


Female 


Adult 


18 


43 


25 


100 


S.australis 


Female 


Adult 


20 


45 


29 


115 


S.australis 


Female 


Subadult 


17 


42 


28 


110 


S.australis 


Female 


Subadult 


16 


45 


24 


95 


S.australis 


Female 


Subadult 


17 


43 


20 


81 


S.australis 


Female 


Subadult 


20 


41 


21 


85 


S.australis 


Male 


Adult 


17 


45 


18 


73 


S.australis 


Male 


Adult 


19 


46 


17 


69 


S.australis 


Male 


Adult 


20 


42 


15 


59 


S.australis 


Male 


Adult 


20 


43 


24 


97 


S.australis 


Male 


Adult 


20 


46 


22 


88 


S.australis 


Male 


Adult 


18 


48 


18 


72 


S.australis 


Male 


Subadult 


18 


38 


23 


92 


S.australis 


Male 


Subadult 


15 


38 


28 


113 


M. minimus 


Female 


Adult 


17 


44 


27 


106 


M.minimus 


Female 


Subadult 


18 


42 


22 


89 


M.minimus 


Male 


Adult 


18 


43 


18 


72 


M.minimus 


Male 


Adult 


20 


44 


21 


84 


M.minimus 


Male 


Adult 


23 


42 


26 


103 



Results 

A total of 90 test-days were completed on 20 wild-caught blossom bats. Both S. 
australis and M. minimus preferred the 15% sugar solution to the 30% solution (p < 
0.001) choosing the 15% solution 54% of the time and the 30% solution 46% of the time 
(Figure 8-3). When presented with the choice between the 15% and 7% solutions, both 
species of bats preferred the 15% solution (69% preference) to the 7% solution (31% 
preference, (p< 0.01). (Figure 8-4) Overall, the 15% solution was preferred to either the 
7% or 30% nectar solution. Together, both species consumed an average of 21.59 + 6.62 



107 

ml (SD) of nectar per night. Average consumption for all bats was 1 17.86 + 39.48% (SD) 
of their body mass per night 



1 30 



I I 20 

a I 15 

•3 1 10 

b , 

£ 5 








01 1 5% concentration 
□ 30% concentration 



■il 
WiMMM 

mm 



S. australis 



M. minimus 



Figure 8-3. Results of 15% or 30% nectar-preference test. 




1 15% concentration 

I 7% f.nnrpntratinn 



S. australis 




Figure 8-4. Results of 1 5% or 7% nectar-preference test. 

Females consumed more nectar solution (p = 0.04) than males (Figure 8-5). Adult 
females consumed an average of 103.25 + 9.71 mis (SD) per bat over four nights, and 
males consumed an average of 79.67 +14.30 mis (SD). Females consumed 160.38 + 
26.57% of their body mass in nectar every night. Males consumed only 108.86 + 40.03% 
of their body mass. 



108 

As a group subadult females consumed an average of 92.00 + 1 1.31 mis (SD) per 
bat over four nights, and subadult males consumed an average of 103.25 + 9.71 mis (SD). 
The largest volume of nectar consumed in a single night was by a subadult male S. 
australis, who consumed 253% of his body mass in nectar (38 ml). 



j- 100 

I 80 

— 

? 60 




I Amount of nectar 
consumed 



Adult Adult male Subadult Subadult 
female female male 

Figure 8-5. Average amount of nectar consumed over four nights for each age class. 

Discussion 
This study found that S. australis and M. minimus prefer the same nectar 
concentrations (15% sucrose) in preference tests, and both species consume them in 
similar volumes. Adult females consumed more than males in both species, and subadults 
consumed more than adults. Adult males consumed the least nectar volume of all groups 
tested. A lack of preference for the highest concentration of 30% nectar indicates that 
blossom bats in Papua New Guinea are not as energy stressed as they are in Australia, 
and that they are not energy-sensitive foragers (Law 1993b) in the more benign climate of 
Papua New Guinea. 



109 

Blossom bats actively seek food sources with a specific nectar content (Law 
1993b, Baker et al. 1998). Bats typically chose the highest concentration of nectar 
available to them to maximize their energy return (Law 1993b, Roces et al. 1993, Law 
1994). However, in Papua New Guinea, the climate is much warmer and constant, and 
bats from lowland Papua New Guinea may not be as energy or food stressed as they are 
in Australia. Ambient temperatures in Papua New Guinea are 25-30°C almost all year, 
and there is equitable rainfall throughout the year (Bonaccorso and McNab 1997). 
Nearby forest and garden habitats provide continuous fruits and flowers during both the 
wet and dry season (Winklemann et al. 2000). In Papua New Guinea, S. austalis has the 
smallest home range of any bat yet studied. This is probably due to the relatively stable 
and abundant food resources, generalized diet of this species, and high availability of 
roost sites. (Winklemann et al. 2000). Home ranges for S. australis in Papua New Guinea 
were 2.7-13.6 ha (Winklemann et al. 2000). Conversely, home ranges were more than 
100 fold larger (12 to 1796 ha) in Australia due to dispersed and less stable food 
resources (Law and Lean 1999). Papua New Guinea and Australia present very different 
energetic stresses and resources to blossom bats living in these two environments. 

Climate can affect the metabolic rates of blossom bats. Populations of S. australis 
and M. minimus from Papua New Guinea responded to progressively cooler temperatures 
with elevated rates of metabolism. S. australis found in the warmer lowland forests of 
Papua New Guinea has a mean basal rate of metabolism 67% of that expected for its 
body mass. In contrast, S. australis from the cooler New Guinea highlands had a mean 
basal rate of metabolism 110% of the expected value for a mammal of its body size 
(Bonaccorso and McNab 1997). Similarly, the metabolism of M minimus in its tropical 



110 

range is only 57% of that expected, and is the lowest relative rate of metabolism known 
for a nectar- feeding bat. Macroglossus minimus from the colder highlands had a basal 
rate that was elevated to 94% of the expected value for its body mass (Bonaccorso and 
McNab 1997, McNab and Bonaccorso 2001). These studies suggest a lack of energy 
stress in the warmer climate and higher metabolic requirements as a result of cooler 
temperatures (McNab and Bonaccorso 2001). Average nectar consumption for all bats in 
this study was 1 1 8% of their body mass per night, whereas in Australia bats consumed 
1 74% of their body mass in nectar nightly (Law 1993b). 

Shallow torpor can be an energy saving mechanism for bats living in the tropics 
(Bartels et al. 1998, Coburn and Geiser 1998). Syconycteris australis entered shallow 
torpor daily to save energy when food and water were withheld (Coburn and Geiser 
1998). The fact that M. minimus has a less well-developed pattern of torpor than S. 
australis may indicate that this species has not had to adapt to the level of energetic stress 
that the more widespread S. australis has adapted to (Bartels et al. 1998). More 
pronounced torpor in subtropical S. australis may be due to low or unpredictable nectar 
availability, or short nights that limit the time available for foraging, and extended time at 
the day roost (Coburn and Geiser 1998). The energy-saving benefits of torpor may not be 
used often with the tropical climate and year round food resources of Papua New Guinea 
(Bonacccorso and McNab 1997). Overall, heterothermic bats seem to be more influenced 
by their size and the climate of their habitat than by their diet or suborder membership 
(Geisner et al. 1996). 

A preference for the 15% solution rather than the 30% solution may not be as 
counterintuitive as first expected. It may actually be more profitable to consume nectars 



Ill 

of lower concentration to elude predation. There is a point beyond which an increase in 
sugar concentration actually decreases the rate of energy yield to the nectar feeder 
(Heyneman 1983). Lower sugar concentrations allow a more rapid fluid uptake, but a 
lower award per unit volume of nectar. Higher nectar concentrations offer a much more 
energy rich reward but limit the uptake of nectar because the rapid increase in nectar 
viscosity sharply reduces the rate of fluid intake (Heyneman 1983, Roces et al. 1993). 
Ambient temperatures influence nectar flow rates. Energy return is higher for a 10% 
sucrose solution at 30°C than for a 22% concentration at 15°C (Heyneman 1983). Hence, 
in Papua New Guinea, where ambient temperatures range from 28-30°C, it is actually 
advantageous for nectar feeding bats to feed on lower concentrations due to the increased 
flow rates at these temperatures. This decreases the time spent at a flower and may result 
in lower predation levels. The preference of S. australis for 15% nectar concentrations 
makes sense in the context of the warmer temperatures of its New Guinean habitat. 
Ambient temperatures place little metabolic stress on the bats, and are near optimum for 
nectar flow. Other bat species consuming nectars with a concentration of 18-21% had a 
balanced daily energy budget (Helverson and Reyer 1984). 

In this study adult females consumed a greater volume of liquid than males in 
both populations, and subadults of both species consumed more than adults. Overall, 
blossom bats in this study consumed 118% of their body mass, but females consumed an 
average of 160% of their body mass daily. Sixty-seven percent of the highest nightly 
consumption volumes (>30 ml) were consumed by females, and they chose the 30% 
solution more often than did males. These results are potentially the result of elevated 
metabolic rates due to the increased energy demands of pregnancy and lactation. For M. 



112 

minimus in Papua New Guinea, breeding is asynchronous and young are born throughout 
the entire year. At any specific locality it is possible to collect females that are nursing 
young, pregnant, or with an early implanted blastocyst or near-term fetus, or non- 
pregnant with stored spermatozoa (Hood and Smith 1989). Adult female S. australis may 
give birth to two young per year (Bonaccorso 1998) at high energetic cost. The energy 
requirements and food intake of pregnant females can be up to 32% higher than non- 
reproducing females (Mattingly and McClure 1982, Thompson 1991). 

Lactation incurs a high energetic cost, being twice as energetically expensive as 
gestation (Racey and Speakman 1987). Barclay (1989) found that as lactation progressed, 
female bats spent more time, and began foraging earlier, to compensate for the energy 
demands of lactation. Post-lactational females may have been over-consuming to reduce 
the energetic effects of recent lactation. Other studies have also found that populations of 
breeding females ate significantly more than mature males of the species (Clark 1980). 

Subadult blossom bats consumed more nectar than adults of both species. Similar 
to females, this may indicate a response to a temporary energetic deficit. Subadults 
conceivably consume more nectar than adults because they are rapidly growing, and their 
need for increased energy has resulted in elevated metabolic demands. Data on post-natal 
growth rates of free-ranging tropical frugivores do not exist (Kunz and Stern 1995, but 
see Kunz and Hood 2001), but as a general rule, small mammals have high growth rates. 
The largest volume of nectar consumed in a single night was by a young male S. 
australis, who consumed 253% of his body mass in one evening. Subadult bats may be 
responding to temporary but intense energetic stress during periods of high growth. Adult 
male fruit bats were overall more stationary, and only moved half as far as females or 



113 

subadults in their nightly foraging flights (Heideman and Heaney 1989). This may 
indicate that adult males are not energetically stressed, as are females and subadults. 
In conclusion, both species of blossom bats preferred 15% nectar solution to 
either 30% or 7% nectar. This may indicate that they are not as energy stressed as 
blossom bats that occur in more temperate and seasonal environments in Australia. The 
second most preferred nectar concentration was the 30% solution, which is energy rich 
and may have been used for bats that were temporarily physiologically stressed, such as 
parous females or growing subadult bats. The 7% solution may offer too little energy and 
not be profitable for bats to consume. Preference for the 15% solution may also be related 
to the nectar flow rates in the warm temperatures of New Guinea. The high temperatures 
and constant food availability of Papua New Guinea seems to result in limited metabolic 
stress to blossom bats that live there. Reproductive females and subadults may be more 
energy stressed than males, but on a temporary basis. 



CHAPTER 9 
CONCLUSION AND CONSERVATION RECOMMENDATIONS 

Testing the Calcium-Constraint Hypothesis 

Robert Barclay (1994,1995) first proposed in his calcium-constraint hypothesis 
that calcium rather than energy may limit reproduction in female bats. Reproductive 
female bats are often calcium deficient following pregnancy and lactation (Kwiecinski et 
al. 1987, Keeler and Studier 1992, Studier et al. 1994a, Studier and Kunz 1995, Bernard 
and Davison 1996). This research experimentally tested the calcium-constraint hypothesis 
using three different species of fruit bats in four different geographic areas. I predicted 
that fruit bats, particularly reproductive females, would seek out and use calcium sources 
to supplement their diet and to relieve calcium deficiencies. 

The most extreme test of the calcium constraint hypothesis included captive 
lactating Malayan flying foxes (P. vampyrus), one of the world's largest flying foxes 
(Chapter 7) with the assumption that their calcium requirements would be very high. 
Malayan flying foxes did not select high calcium foods in this study. Instead, they 
rejected high-calcium foods and preferred fruits high in sugar. P. vampyrus did not 
appear to meet its calcium needs and showed elevated absorption levels, a potential sign 
of calcium deficiency. 

Tongan fruit bats (P. tonganus) in American Samoa also rejected calcium-rich 
fruits and instead preferred high-energy, sugar-rich fruits in fruit choice experiments 
(Chapter 3). They also preferred agricultural areas while foraging throughout the island 
landscape (Chapter 6). Mineral metabolism experiments indicate that wild Tongan fruit 

114 



115 

bats did not meet many of their mineral requirements while consuming a diet consisting 
primarily of agricultural fruits (Chapter 5). P. tonganus consumed only 1/8 of the 
recommended calcium requirement on their chosen diet. In addition, apparent absorption 
values for the Tongan fruit bat population were higher for lactating female P. vampyrus, 
potentially indicating mineral deficiency throughout the Tongan fruit bat population. 

Both Malayan and Tongan fruit bats did not seem to be selecting foods in a 
manner that maximized their calcium intake. Their preference for fruits high in sugar may 
instead indicate an energy constraint. To examine bat consumption to maximize energy, I 
utilized two of the smallest members of Old World fruit bats, the blossom bats 
Syconycteris australis and Macroglossus minimus (Law 1992, 1993b, 1994). Neither 
species preferred the most concentrated energy source and instead preferred the 
intermediate 15% sugar solution. However, females and young adults consumed the 
largest volumes of nectar, possibly indicating energy stress among those groups due to 
reproduction and rapid growth. 
Mineral Compensation 

Tongan fruit bats did seek out and consume calcium-rich resources on a limited 
basis. The strongest indicator that P. tonganus may be compensating for their calcium- 
poor diet was the frequency of leaf-consumption by captive bats (Chapter 4). Most of the 
Tongan fruit bats in captivity (83%) ate leaves, which had the potential to increase their 
calcium consumption by 46%. Although not high in sugar or energy nor especially 
palatable, leaves represent a readily available and consumed calcium-rich source of 
minerals (Kunz and Ingalls 1994, Kunz and Diaz 1995, Nelson et al. 2000b). 

The second indicator that fruit bats may supplement their diet with additional 
calcium was seen in female use of calcium blocks (Chapter 3). Three times as many 



116 

females as males used the calcium blocks in the feeding experiments, with 100% use and 
the highest frequency of use among reproductive females, although the sample size was 
very small. The members of the group that were most vulnerable to calcium deficiency, 
including reproductive females and rapidly growing young males, used the calcium 
blocks most often. This may indicate that calcium is consumed to relieve deficiencies 
among these individuals. 

Fruit bats may also compensate for a lack of calcium in their diet by their unique 
manner of feeding, which possibly results in greater calcium availability (Chapter 5). By 
consuming mostly fruit juices, minerals like calcium maybe swallowed in solution rather 
than bound to fiber, and may result in a higher degree of bioavailability (Ammerman 
1995). A specialized gastrointestinal tract may also help to increase calcium absorption. 
Tongan and Malayan Fruit Bat Dietary Choice 

Tongan and Malayan females bats did not appear to consume a nutritionally 
adequate diet. Their diet was marginal for most of the macronutrients examined in this 
study and especially poor in calcium. If and how Tongan and Malayan females are able to 
rebuild their skeletons following pregnancy and lactation, and support continued 
population growth remains unknown. Because female reproduction issues are critical to 
the health of the population and the continuation of the species through time, the food 
choices made by both P. tonganus and P. vampyrus seem maladaptive. 

The fruits chosen by these two species, and the mineral intake that results, may 
not be typical of other fruit bat species. The Samoan fruit bat, P. samoensis, co-occurs 
with P. tonganus in American Samoa, but feeds primarily in native forest, consuming a 
wide variety of calcium-rich native fruits including figs (Trail 1994, Banack 1996). 
Similarly, A.jubatus, which co-roosts with P. vampyrus, consumes a diet rich in figs and 



117 

native fruits (Stier and Mildenstein 2001). Most fruit bat species consume fruits from the 
native rainforest (Palmer and Woinarski 1999, Palmer et al. 2000, Brooke 2001) or even 
preferentially consume calcium-rich foods (Barclay 2002). Therefore, the results of this 
research may be typical of fruit bat species that use agricultural areas, but may be atypical 
of bats that utilize native forest for the majority of their feeding. It is unknown what 
Tongan fruit bats ate before the arrival of Polynesians 3,000 years ago and the planting of 
agricultural fruits (Kirch and Hunt 1997). Tongan fruit bats may have previously foraged 
in a manner more consistent with that of Samoan fruit bats and consumed a higher quality 
diet. 

Summary 

This research attempted to distinguish if reproduction in females was constrained 
more by energy or calcium (Barclay 1994, 1995). The results of this research indicate that 
bats forage to maximize their energy intake by preferring sugar-rich fruits. However, in 
times of physiological stress such as pregnancy and lactation, or during rapid growth by 
young bats, fruit bats may exhibit compensatory foraging for deficient minerals such as 
calcium. Use of leaves and calcium blocks may highlight a drive to relieve mineral 
deficiencies by consuming concentrated mineral sources. 

Foraging habits of reproductive females were of special consideration in this 
research. However, generalizations derived from their choices are limited because of the 
small sample size of captured reproductive females. Much work remains to be done in the 
area of nutritional ecology to further distinguish what motivates food choice among Old 
World flying foxes and how reproductive females meet their nutritional requirements. 



118 

Conservation Recommendations 
Major Threats to Fruit Bats 

Human activities constitute the greatest threat to fruit bat populations. Fruit bats 
can be harmed either directly by hunting at the local or commercial level, or indirectly by 
large-scale destruction of habitat and roost disturbance. Fruit bats have long been 
consumed by local people (Wodzicki and Felton 1980, Cheke and Dahl 1981, Heaney 
and Heideman 1987, Rainey 1990). Hunting continues at both the local and commercial 
level, often as part of a luxury food trade (Racey 1979, Wiles and Payne 1986, Fujita and 
Tuttle 1991) or by commercial fruit growing operations, where fruit bats are considered 
pests (Loebel and Sanewski 1987). 

Habitat loss and forest destruction are now considered the major factors 
contributing to fruit bat population decline (Cheke and Dahl 1981, Fujita and Tuttle 1991, 
Mickleburgh et al. 1992, Pierson and Rainey 1992). Loss of forest results in loss of both 
critical food and roosting resources as native forest is logged to make way for residential 
and/or commercial development and agriculture (Mickleburgh et al. 1992). Many fruit 
bat species roost deep in native forest, and disturbance to the roost, either accidental or 
intentional, can result in abandonment that can have population level effects during the 
maternity season (Wiles 1987b, Brooke 1998). Natural catastrophes such as periodic and 
destructive hurricanes can also lead to dramatic population declines in fruit bat 
populations (Craig et al. 1994b). 
The National Park of American Samoa 

The National Park of American Samoa on Tutuila, Ofu, and Olosega, American 
Samoa, is important regionally as a refuge for Tongan and Samoan fruit bats that reside 
there (Brooke 1998). The National Park provides primary forest and roosting locations 



119 

that are secluded from human disturbance and protected from hunting. Native forests are 
important to Samoan fruit bats for foraging and roosting (Banack 1996, Brooke 2001). 
Tongan fruit bats also favor isolated and inaccessible roosts in native rainforest (Brooke 
1998). There are at least four traditional Tongan bat roost sites within the National Park 
on Tutuila that are currently in use (Brooke 1998) Additional roosts may be established 
within the National Park as native forest areas continue to dwindle as a result of the rapid 
human population growth (Brooke 1998, Craig et al. 2000). Although a substantial 
amount of forest is contained within the National Park, it may not be adequate to 
maintain viable populations of bats if forest clearance continues at the current rate 
(Rainey 1998). Any opportunity to enlarge the current boundaries of the National Park, 
particularly along the north coast of Tutuila, should be seized (Brooke 1998). 

On a larger scale, long-term and landscape-level conservation plans must be 
enacted, and native forests set aside as conservation areas to protect current and future 
populations of fruit bats within oceanic islands in the South Pacific. A collection of 
conservation areas can act as refuges for bat populations and can serve to repopulate 
areas following hurricanes and other catastrophes. 
Reducing Hunting Pressure and Bat Education Programs 

A ten-year hunting ban, in effect since 1992, has reduced hunting in American 
Samoa (Craig and Syron 1992, Brooke 2001), but hunting of fruit bats continues (S. 
Nelson, pers. obs.). The hunting ban was enacted to address declines in fruit bat 
populations resulting from the combination of hurricane damage to the forest following 
three successive hurricanes, and the opportunistic hunting of bats immediately 
afterwards. Together, these factors contributed to an 80-90% decline in populations of 



120 

fruit bats (Craig et al. 1994b). When questioned, American Samoans were either largely 
uninformed of the ban, or killed and consumed bats despite it. Therefore, I suggest 
aggressive education programs in schools and villages explaining the reason for the 
hunting ban. I also suggest using the widely read local newspaper for articles explaining 
the ban, and to publish reminders of the hunting ban weekly or monthly. My experience 
talking to schoolchildren indicated that residents of the most remote villages on Tutuila 
were least aware of the role of bats in forest ecosystems and also engaged in the most bat 
hunting. Hence, these areas should be a priority for education programs on the island. 
Although discussed for Tutuila, American Samoa, these recommendations can be widely 
applied to other Pacific island ecosystems where bats are hunted unsustainable 

The fate of the native forest and the fruit bat populations that reside within them 
are interconnected. Native forest provides fruits and roosts to fruit bats, and fruit bats are 
crucial to the maintenance and integrity of native forests. Due to the paucity of alternative 
vertebrate pollinators and seed dispersers on isolated oceanic islands, fruit bat species are 
often considered keystone species (Cox et al. 1991, Rainey et al. 1995, Banack 1998). 
Thus, declines or extinctions among fruit bat populations may have long-term impacts on 
forest regeneration (Bonaccorso and Humphrey 1978, Thomas 1982, Rainey et al. 1995) 
and cascade effects on other species that reside in native forest (Rainey 1998). A loss or 
significant decline in bat populations may ultimately affect community structure, 
biodiversity and ecosystem function in simple systems such as oceanic islands. 



LIST OF REFERENCES 

Allen, ME., Oftedal, O.T., 1996. Essential nutrients in mammalian diets. In: Kleiman, 
D.G., Allen, M.E., Thompson, K.V., Lumpkin, S. (Eds.), Wild Animals in 
Captivity, Principles and Techniques. The University of Chicago Press. Chicago. 

Amerson, A.B. Jr., Whistler, W.A., Schwaner, T.D., 1982. Wildlife and wildlife habitat 
of American Samoa II, accounts of flora and fauna. U.S. Department of the 
Interior, Fish and Wildlife Service Washington, D.C. 

Ammerman, C.B., 1995. Methods for estimation of mineral bioavailability. In: 

Ammerman, C.B., Baker, D.H., Lewis, A.J. (Eds.), Bioavailability of Nutrients 
for Animals: Amino acids, Minerals and Vitamins. Academic Press, Inc. New 
York. 

Anderson, K., 1912. Catalogue of the Chiroptera in the collection of the British Museum. 
2nd ed. British Museum (Natural History), London. 

Anthony, E.L.P., 1988. Age Determination in bats. In: Kunz, T.H. (Ed.), Ecological and 
Behavioral Methods for the Study of Bats. Smithsonian Institution Press, 
Washington, D.C. 

Arnaud, CD., Sanchez, S.D., 1996. Calcium and phosphorous. In: Ziegler, E.E., Filer, 
L.J., Jr., (Eds.), Present Knowledge in Nutrition. 7 th ed. ELSI Press. Washington 
D.C. 

Arnold, G.W., 1964. Some principles in the investigation of selective grazing. 
Proceedings of the Australian Society of Animal Production. 5, 258. 

Augspurger C. K., 1983. Seed dispersal in the tropical tree, Platypodium elegans, and the 
escape of its seedlings from fungal pathogens. Journal of Ecology 71, 759-771. 

Baillie, J., Groombridge, B., 1996. 1996 IUCN Red List of Threatened Animals. IUCN, 
Gland, Switzerland. 

Baker, H. G., Baker, I., Hodges, S. A., 1998. Sugar composition of nectars and fruits 

consumed by birds and bats in the tropics and subtropics. Biotropica 30, 559-586. 

Banack, S.A., 1996. Flying foxes, genus Pteropus, in the Samoan Islands: interactions 
with forest communities. PhD thesis, University of California, Berkeley. 

Banack, S.A., 1998. Diet selection and resource use by flying foxes (genus Pteropus). 
Ecology 79, 1949-1967. 

Barclay, R.M.R., 1989. The effect of reproductive condition on the foraging behavior of 
female hoary bats, Lasiurus cinereus. Behavioral Ecology and Sociobiology 24, 
31-37. 



121 



122 



Barclay, R.M.R., 1994. Constraints on reproduction by flying vertebrates: energy and 
calcium. American Naturalist. 144, 1021-1031. 

Barclay, R.M.R. 1995. Does energy or calcium availability constrain reproduction in 
bats? Symposium of the Zoological Society of London. 67, 245-258. 

Barclay, R.M.R., 2002. Do plants pollinated by flying fox bats (Megachiroptera) provide 
an extra calcium reward in their nectar? Biotropica 34, 168-171. 

Bartels, W., Law, B.S., Geiser, F., 1998. Daily torpor and energetics in a tropical 

mammal, the northern blossom-bat Macroglossus minimus (Megachiroptera). 
Journal of Comparative Physiology B-Biochemical, Systematic, and 
Environmental Physiology 168, 233-239. 

Bastian, S.T., Tanaka, K., Anunciado, R.V.P., Natural, N.G., Sumalde, A.C., 

Namikawa.T., 2001. Phylogenetic relationships among megachiroptera species 
from the two major islands of the Phillipines, deduced from DNA sequences of 
the cytochrome b gene. Canadian Journal of Zoology 79, 1671-1677. 

Batzli, G.O., 1986. Nutritional ecology of the California vole: effects of food quality on 
reproduction. Ecology 67, 406-412. 

Batzli, GO., Pitelka, FA., 1983. Nutritional ecology of microtine rodents- food habits of 
lemmings near Barrow, Alaska. Journal of Mammalogy. 64, 648-655. 

Belovsky, G.E., Jordan, PA., 1981. Sodium dynamics and adaptations of a moose 
population. Journal of Mammalogy 62, 613-621. 

Bernard, R.T.F., Davison, A., 1996. Does calcium constrain reproductive activity in 
insectivorous bats? Some empirical evidence for Schreibers' long-fingered bat 
{Miniopterus schreibersii). South African Journal of Zoology 31, 218-220. 

Bolten, A., Feinsinger, B., Baker, H.G., Baker, I., 1979. On the calculation of sugar 
concentration in flower nectar. Oecologia 41, 301-304. 

Bonaccorso, F.J., 1998. Bats of Papau New Guinea. Conservation International, 
Washington D.C. 

Bonaccorso, F.J., Humphrey, S.R., 1978. Fruit bat niche dynamics: their role in 

maintaining tropical forest diversity. Pp. 169-183 in Chadwick, A.C., Sutton, S.L., 
Eds. Tropical Rain-Forest: the Leeds Symposium. Leeds Philosophical and 
Literary Society, Leeds. 

Bonaccorso, F.J., McNab, B.K., 1997. Plasticity of energetics in blossom bats 

(Pteropodidae): impact on distribution. Journal of Mammalogy 78, 1073-1088. 

Brautigam, A., 1992. Conservation of Pacific island flying foxes and the convention on 
international trade in endangered species of wild fauna and flora. In: Wilson,D.E., 
Graham, G.L. (Eds.), Pacific Island Flying Foxes: Proceedings of an International 
Conservation Conference. U.S. Fish and Wildlife Service. Washington, D.C. 

Brautigam, A., Elmqvist, T., 1990. Conserving Pacific island flying foxes. Oryx 24, 81- 
89. 



123 



Brine, C.L., Johnston, F.A., 1955. Endogenous calcium in the feces of the adult man and 
the amount of calcium absorbed from food. American Journal of Clinical 
Nutrition 3, 418. 

Brink, E.J., Dekker, P.R., Vanberesteijn, E.C.H., Beynen, A.C., 1992. Bioavailability of 
magnesium and calcium from cows milk and soybean beverage in rats. British 
Journal of Nutrition 68, 271-282. 

Brommage, R., 1989. Measurement of calcium and phosphorous fluxes during lactation 
in the rat. Journal of Nutrition 119, 428-438. 

Brommage, R., DeLuca, H., 1984. Self-selection of a high calcium diet by vitamin D- 
deficient lactating rats increases food consumption and milk production. Journal 
of Nutrition 114, 1377-1385. 

Bronner, F., 1987. Intestinal calcium-absorption mechanisms and applications. Journal of 
Nutrition 117, 1347-1352. 

Bronner, F., 1998. Calcium absorption- a paradigm for mineral absorption. Journal of 
Nutrition 128, 917-920. 

Bronner, F., Pansu, D., 1999. Nutritional aspects of calcium absorption. Journal of 
Nutrition 129, 9-12. 

Bronner, F., Salle, B.L., Putet, G., Rigo, J., Senterre, J., 1992. Net calcium absorption in 
premature infants: results of 103 metabolic balance studies. American Journal of 
Clinical Nutrition 56, 1037-1044. 

Bronstein, J.L., 1995. The plant-pollinator landscape. In: Hansson, L., Fahrig, L., 

Merriam, G., (Eds.). Mosaic Landscapes and Ecological Processes. Chapman and 
Hall. London. 

Brooke, A.P., 1998. Biology of the flying foxes in American Samoa: Pteropus samoensis 
and Pteropus tonganus; a report to the National Park of American Samoa. Pago 
Pago, American Samoa. 

Brooke, A. P., 2001. Population status and behaviours of the Samoan flying fox {Pteropus 
samoensis) on Tutuila Island, American Samoa. Journal of Zoology, London 254, 
309-319. 

Brooke, A.P., Solek, C, Tualauleilei, A., 2000. Roosting behavior of colonial and 

solitary flying foxes in American Samoa (Chiroptera: Pteropodidae). Biotropica 
32, 338-350. 

Cheke, A.S., Dahl, J.F., 1981. The status of bats on western Indian Ocean islands with 
special reference to Pteropus. Mammalia 45, 205-238. 

Christopherson, E., 1935. Flowering plants of Samoa. Bernice Pauahi Bishop Museum 
Bulletin 128, 1-221. 

Clark, D. A., 1980. Age and sex dependent foraging strategies of a small mammalian 
omnivore. Journal of Animal Ecology 49, 549-563. 



124 



Cobum, D.K., Geiser, F., 1998. Seasonal changes in energetics and torpor patterns in the 
subtropical blossom-bat Syconycteris australis (Megachiroptera). Oecologia 113, 

467-473. 

Coen, C, 2002. Comparative nutritional ecology of two genera of vampire bats: 

Desmodus rotundus and Diaemus youngi. PhD thesis, Cornell University, Ithaca. 

Cole, F.R., Batzli, G.O., 1979. Nutrition and population dynamics of the prairie vole 

Microtus ochrogaster, in central Illinois. Journal of Animal Ecology 48, 455-470. 

Cole, T.G., Whitesell, CD., Whistler, W.A., McKay, N., Ambacher, A.H., 

1988. Vegetation survey and forest inventory, American Samoa. USDA Forest 
Service Report, Report PSW-25. Pacific SW Forest and Range Experiment 
Station, Berkeley CA. 

Cook, S., Singidan, R., Thornton, I.W.B., 2001. Colonization of an island volcano, Long 
Island, Papua New Guinea, and an emergent island, Motmot, in its caldera lake. 
IV. Colonization by non-avian vertebrates. Journal of Biogeography 28, 1353- 
1363. 

Coppock, C.E., Everett, R.W., Merrill, W.G., 1972. Effect of ration on free choice 

consumption of calcium-phosphorus supplements in dairy cattle. Journal of Dairy 
Science 55, 245-256. 

Corbet, G.B., Hill, J.E., 1992. Mammals of the Indomalayan Region: A Systematic 
Review. Oxford University Press, New York. 

Courdray, C, Bellanger, J., CastigliaDelavaud, C, Remsey, C, Vermorel, M., 
Rayssignuier, Y., 1997. Effect of soluble or partly soluble dietary fiber 
supplementation on absorption and balance of calcium, magnesium, iron and zinc 
in healthy young men. European Journal of Clinical Nutrition 51, 375-380. 

Courts, S.E., 1998. Nutritional studies of Livingstone's fruit bat Pteropus livingstonii: 

feeding behaviour, diet evaluation, and modification. Jersey Wildlife Preservation 
Trust. 

Courts, S.E., Feistner, A.T.C., 2000. Nutrition of Old World fruit bats in captivity: diet 

studies of Livingstone's fruit bats Pteropus livingstonii. In: Nijboer, J., Hatt, J.M., 
Kaumanns, W., Beijen, A., Ganslober, U., (Eds.), Zoo Animal Nutrition. Filander 
Verlag, Germany. 

Cox, P. A., Elmqvist, T., Pierson, E.D., Rainey, W.E., 1991. Flying foxes as strong 
interactors in South Pacific island ecosystems: a conservation hypothesis. 
Conservation Biology 5, 448-454. 

Cox, P. A., Elmqvist, T., Pierson, E.D., Rainey, W.E. 1992. Flying foxes as pollinators 
and seed dispersers in Pacific island ecosystems. In: Wilson, D.E., Graham, G.L. 
(Eds.), Pacific island flying foxes: Proceedings of an International Conservation 
Conference. U.S. Fish and Wildlife Service. Washington, D.C. 

Cox, P.A., Elmqvist, T., 2000. Pollinator extinction in the Pacific Islands. Conservation 
Biology 14, 1237-1239. 



125 

Craig, P., Morrell, T.E., So'oto, K., 1994a. Subsistence harvest of birds, fruit bats, and 
other game in American Samoa, 1990-1991. Pacific Science 48, 344-352. 

Craig, P., Saucerman, S., Wiegman, S., 2000. The central South Pacific Ocean (American 
Samoa) In: Sheppard, C. (Ed.), Seas at the Millennium: An Environmental 
Evaluation. Elsevier Science, Ltd. Oxford, U.K. 

Craig, P., Syron, W., 1992. Fruit bats in American Samoa: their status and future. In: 

Wilson, D.E., Graham, G.L. (Eds.), Pacific Island Flying Foxes: Proceedings of 
an International Conservation Conference. U. S. Fish and Wildlife Service. 
Washington, D.C. 

Craig, P., Trail, P., Morrell, T.E., 1994b. The decline of fruit bats in American Samoa 
due to hurricanes and overhunting. Biological Conservation 69, 261-266. 

Dasilva, G.L., 1994. Diet of Colobus polykomos on Tiwai Island: selection of food in 

relation to its seasonal abundance and nutritional quality. International Journal of 
Primatology 15, 655-678. 

Delguidice, G.D., Mech, L.D., Seal, U.S., 1990. Effects of winter undernutrition on body 
composition and physiological profiles of white-tailed deer. Journal of Wildlife 
Management 54, 539-550. 

Dempsey, J.L., 1999. Advances in fruit bat nutrition. In: Fowler, M.E., Miller, R.E., 
(Eds.), Zoo and Wild Animal Nutrition: Current Therapy. W.B. Saunders Co., 
Philadelphia. 

DeSantiago, S., Alonso, L., Halhali, A., 1999. Maternal calcium metabolism during 
lactation. Revista de Investigacion Clinica. 51, 309-314. 

Dierenfeld, E.S., McCann, CM., 1999. Nutrient composition of selected plant species 
consumed by semi free-ranging lion-tailed macaques (Macaco, selenus) and ring- 
tailed lemurs (Lemur catta) on St. Catherine's Island, Georgia, U.S.A. Zoo 
Biology 18, 481-494. 

Dierenfeld, E.S., Seyjagat, J., 2000a. Plasma fat soluble vitamin and mineral 

concentrations in relation to diet in captive Pteropodid bats. Journal of Zoo and 
Wildlife Medicine 31, 315-321. 

Dierenfeld, E.S., Seyjagat, J., 2000b. Intake and diet digestibility in three species of 

captive Pteropodid bats. In: Nijboer, J., Hatt, J.M., Kaumanns, W., Beynen, A., 
Ganslober, U, (Eds.), Zoo Animal Nutrition. Filander Verlag, Germany. 

Dominique, C.P., 1991. Feeding strategy and activity budget of the frugivorous bat 

Carollia perspicillata (Chiroptera, Phyllostomidae) in French Guiana. Journal of 
Tropical Ecology 7, 243-256. 

Drew, R.A.I., 1988. Amino-acid increases in fruit infested by fruit-flies of the family 
Tephritidae. Zoological Journal of the Linnean Society 93, 107-112. 

Dudley, R., Vermeij, G.J., 1992. Do the power requirements of flapping flight constrain 
folivory in flying animals? Functional Ecology 6, 101-104. 



126 



Dudley, R., Vermeij, G.J., 1994. Energetic constraints of folivory: leaf fractionation by 
frugivorous bats. Functional Ecology 8, 668. 

Duflos, C, Bellaton, C, Pansu, D., Bronner, F., 1995. Calcium solubility, intestinal 
sojourn time and paracellular permeability codetermine passive calcium 
absorption in rats. Journal of Nutrition 125, 2348-2355. 

Eby, P., 1991. Seasonal movements of grey-headed flying foxes, Pteropus poliocephalus 
(Chiroptera: Pteropodidae), from two maternity camps in Northern New South 
Wales. Wildlife Research 18, 547-549. 

Elangovan, V., Marimuthu, G., Kunz, T.H., 2001. Temporal patterns of resource use by 
the short-nosed fruit bat, Cynopterus sphinx (Megachiroptera: Chiroptera). 
Journal of Mammalogy 82, 161-165. 

Elmqvist, T., Cox, P. A., Rainey, W.E., Pierson, E.D., 1992. Restricted pollination on 
oceanic islands- pollination of Ceiba pentandra by flying foxes in Samoa. 
Biotropica 24, 15-23. 

Elmqvist, T., Rainey, W.E., Pierson, E.D., Cox, P.A., 1994. Effects of tropical cyclones 
Ofa and Val on the structure of a Samoan lowland rain forest. Biotropica 26, 384- 
391. 

Erickson, W.P., McDonald, T.L., Gerow, K.G, Howlin, S., Kern, J.W., 2001. Statistical 
issues in resource selection studies with radio-marked animals. In: Millspaugh, 
J.J., Marzluff, J.M., (Eds.), Radio tracking and animal populations. Academic 
Press, New York. 

Faber, W.E., Pehrson, A., Jordan, PA., 1993. Seasonal use of salt blocks by mountain 
hares in Sweden. Journal of Wildlife Management 57, 842-846. 

Fairweather-Tait, S.J., Wright, J. A., 1991. Small intestinal transit time and iron 
absorption. Nutrition Research 11, 1465. 

Fascione, N., 1995. Fruit bat husbandry manual. Published by the Lubee Foundation, 
Inc., and American Association for Zoos and Aquaria. Gainesville, FL. 

Flannery, T.F., 1995a. Mammals of New Guinea (2 nd ed.) Australian Museum/Reed 
Books, New South Wales. 

Flannery, T.F., 1995b. Mammals of the Southwest Pacific and Moluccan Islands. Cornell 
University Press, New York. 

Fleming, T.H., 1988. The Short-Tailed Fruit Bat. University of Chicago Press, Chicago. 

Freifeld, H.B., 1998. Temporal and spatial variation in forest birds on Tutuila Island, 
American Samoa. PhD thesis. University of Oregon, Eugene. 

Freifeld, H.B., 1999. Habitat relationships of forest birds on Tutuila Island, American 
Samoa. Journal of Biogeography 26, 1191-1213. 

Fujita, M.S., Tuttle, M.D., 1991. Flying foxes (Chiroptera: Pteropodidae): threatened 
animals of key ecological and economic importance. Conservation Biology 5 
455-463. 



127 



Funakoshi, K., Watanabe, H., Kunisaki, T., 1993. Feeding ecology of the northern 

Ryukyu fruit bat, Pteropus dasymallus dasymallus, in a warm-temperate region. 
Journal of the Zoological Society of London. 230, 221-230. 

Galindo, A., Atkinson, K., Gross, T., 1995. Reproduction survey of Pteropodids in North 
American zoos and wild animal parks. Published by the Lubee Foundation, Inc. 
for the AZA Bat Taxon Advisory Group. Gainesville, FL. 

Garel, J.M., 1987. Hormonal control of calcium metabolism during the reproductive 
cycle in mammals. Physiological Reviews 67, 1-66. 

Garigal, K., Cushman, S., Stafford, S., 2000. Multivariate statistics for Wildlife and 
Ecology research. Springer- Verlag New York, Inc. New York. 

Geiser, F., Coburn, D.K., 1999. Field metabolic rates and water uptake in the blossom-bat 
Syconycteris australis (Megachiroptera). Journal of Comparative Physiology B- 
Biochemical Systemic and Environmental Physiology. 169, 133-138. 

Geiser, F., Coburn, D.K., Kortner, G, Law, B.S., 1996. Thermoregulation, energy 

metabolism, and torpor in blossom-bats, Syconycteris australis (Megachiroptera). 
Journal of Zoology 239, 583-590. 

Goodwin, R.E., 1979. The bats of Timor. Bulletin of the American Museum of Natural 
History. 163, 75-122. 

Gould, E., 1977. Foraging behavior of Pteropus vampyrus on the flowers of Durio 
zibethinus. Malaysian Nature Journal 30, 53-57. 

Gould, E., 1978. Foraging behavior of Malaysian nectar feeding bats. Biotropica 10, 184- 
187. 

Grant, G.S., Banack, S.A., Trail, P., 1994. Decline of the sheath-tailed bat Emballonura 
semicaudata (Chiroptera: Emballonuridae) on American Samoa. Micronesica 27, 
133-137. 

Green, P.E., Srinivasan, V., 1978. Conjoint analysis in consumer research: issues and 
outlook. Journal of Consumer Research 5, 103-123. 

Hair, J.F.Jr., Anderson, R.E., Tatham, R.L., Black, W.C., 1998. Multivariate data 
analysis. 5 th edition. Prentice Hall, NJ. 

Hall, L., Richards, G, 2000. Flying foxes, fruit and blossom bats of Australia. Krieger 
Publishing Co., New South Wales. 

Harris, B.J., Baker, H.G, 1959. Pollination of flowers bats in Ghana. The Nigerian Field 
24, 151-159. 

Harris, W.D., Popat, P., 1954. Determination of the phosphorous content of lipids. 
American Oil Chemical Society Journal 31, 124. 

Havera, S.P., 1978. Nutritional supplement feeding, and body composition, of the fox 

squirrel, Sciurus niger, in Central Illinois. PhD Thesis. The University of Illinois, 
Urbana, IL. 

Hayssen, V., Kunz, T.H., 1996. Allometry of liter mass in bats: maternal size, wing 
morphology, and phenology. Journal of Mammalogy 77, 476-490. 



Heaney, L.R., Heideman, P.D., 1987. Philippine fruit bats: endangered and extinct. Bats 

5,3-5. 

Heard, D.J., Whittier, D.A., 1997. Hematologic and plasma biochemical reference values 
for three flying fox species (Pteropus sp.). Journal of Zoo and Wildlife Medicine 
28, 464-470. 

Heideman, P.D., 1995. Synchrony and seasonality of reproduction in tropical bats. In: 
Racey, P. A., Swift, S.M., (Eds.), Ecology, Evolution and Behaviour of Bats. 
Clarendon Press, Oxford, U.K. 

Heideman P.D., Heaney, L.R., 1989. Population biology and estimates of abundance in 
fruit bats (Megachiroptera, Pteropodidae). Journal of Zoology, London 218, 565- 
586. 

Helverson, O., Reyer, H., 1984. Nectar intake and energy expenditure in a flower 
visiting bat. Oecologia 63, 178-184. 

Herbst, L.H., 1986. The role of nitrogen from fruit pulp in the nutrition of the frugivorous 
bat Carol Ha perspicillata. Biotropica 18, 39-44. 

Herrera, CM., 1987. Vertebrate-dispersed plants in the Iberian peninsula- a study of fruit 
characteristics. Ecological Monographs 57, 305-331. 

Hevesey, G., 1948. Radioactive Indicators. Wiley Interscience, New York. 

Heyneman, A. J., 1983. Optimal sugar concentrations of floral nectars- dependence on 
sugar intake efficiency and foraging costs. Oecologia 60, 198-213. 

Hjerpe, J., Hedenas, H., Elmqvist, T., 2001. Tropical rain forest recovery from cyclone 
damage and fire in Samoa. Biotropica 33, 249-259. 

Holick, M.F., 2002b. Evolution and function of vitamin D. Anti-Cancer Drugs 13, 5-12. 

Holick, M.F., 2002a. Too little vitamin D in premenapausal women: why should we care? 
American Journal of Clinical Nutrition 76, 3-4. 

Holl, S.A., Bleich, V.C., 1987. Mineral lick use by mountain sheep in the San-Gabriel 
Mountains, California. Journal of Wildlife Management 51, 383-385. 

Hollander, M., Wolfe, D.A., 1999. Nonparametric Statistical Methods. 2 nd Ed. John 
Wiley and Sons, Inc. New York. 

Hood, C.S., Smith, J.D., 1989. Sperm storage in a tropical nectar feeding bat, 

Macroglossus minimus (Pteropodidae). Journal of Mammalogy 70, 404-406. 

Hood, W.R., Kunz, T.H., Oftedal, O.T., Iverson, S.J., LeBlanc, D., Seyjagat, J., 2001. 
Interspecific and intraspecific variation in proximate, mineral, and fatty acid 
composition of milk in old world fruit bats (Chiroptera: Pteropodidae). 
Physiological and Biochemical Zoology. 74, 134-146. 

Hoying, K.M., Kunz, T.H., 1998. Variation in size at birth and post-natal growth in the 
insectivorous bat Pipistrellus subflavus (Chiroptera: Vespertilionidae). Journal of 
Zoology 245, 15-27. 



129 



Hunt, T.L., Kirch, P.V., 1997. The historical ecology of Ofu Island, American Samoa, 

3000 B.P. to present. In: Kirch, P.V., Hunt, T.L., (Eds.), Historical Ecology of the 
Pacific Islands. Yale University Press. New Haven, CT. 

Hurley, J.S., Keen, C.L., 1987. Manganese. In: Mertz, W., (Ed.), Trace Elements in 
Human and Animal Nutrition. 5 th ed. Vol. 1 . Academic Press, New York. 

Hurwitz, S., 1996. Homeostatic control of plasma calcium concentration. Critical 
Reviews in Biochemistry and Molecular Biology 31, 41-100. 

Ingle, N.R., Heaney, L.R., 1992. A key to the bats of the Phillipine Islands. Fieldiana: 
Zoology New Series 69, 1-44. 

Itino, T., Kato, M., Hotta, M., 1991. Pollination ecology of two wild bananas, Musa 
acuminata subsp. halabensis and M. salaccensis: chiropterophily and 
ornithophily. Biotropica 23, 151-158. 

Jantzen, D.H., Miller, G.A., Hackforth-Jones, J., Pond, CM., Hooper, K., James, 
D.P.,1976. Two Costa Rican bat-generated seed shadows of Andira inermis 
(Swartz) HBK, Leguminosae. Ecology 57, 1068-1076. 

Jones, R.L., Hanson, H.C., 1985. Mineral Licks, Geophagy, and Biochemistry of North 
American Ungulates. Iowa State University Press, Ames. 

Josephs, H.W., 1958. Absorption of iron as a problem in human physiology- a critical 
review. Blood 13, 1-54. 

Jungbloed, A.W., Kemme, PA., 1990. Effect of pelleting mixed feeds on phytase activity 
and apparent absorbability of phosphorous and calcium in pigs. Animal Feed 
Science and Technology 28, 233-242. 

Kear, D., Wood, B.L., 1959. The geology and hydrology of Western Samoa. NewZealand 
Geological Survey Bulletin 63, 1-90. 

Keeler, J.O., Studier, E.H., 1992. Nutrition in pregnant big brown bats (Eptesicus fuscus) 
feeding on June beetles. Journal of Mammalogy 72, 426-430. 

Kenward, R.E., 2001. A Manual for Wildlife Radio Tagging. Academic Press, New 
York. 

Kimball, B.A., Nolte, D.L., Engeman, R.M., Johnston, J. J., Stermitz, F.R., 1998. 

Chemically mediated foraging preference of black bears {Ursus americanus). 
Journal of Mammalogy 79, 448-456. 

Kirch, P. V., Flenley, J.R., Steadman, D.W., Lamont, F., Dawson, S., 1992. Prehistoric 

human impacts on an island ecosystem, Mangaia, central Polynesia. Research and 
Exploration8, 166-179. 

Kirch, P.V., Hunt, T.L., 1993. The To'aga site: Three Millennia of Polynesian 

Occupation in the Manu'a Islands, American Samoa. Archaelogical Research 
Facility, University of California, Berkeley 

Kirsch, J.A., Flannery, T.F., Springer, M.S., Lapointe, F., 1995. Phylogeny of the 
Pteropodidae (Mammalia: Chiroptera) based on DNA hybridization, with 
evidence for bat monophyly. Australian Journal of Zoology 43, 395-428. 



130 



Klaus, G., Klaus-Hugi, C, Schmid, B., 1998. Geophagy by large mammals at natural 

licks in the rain forest of Dzanga National Park, Central African Republic. Journal 
of Tropical Ecology 14, 829-839. 

Koo, W.W.K., Tsang, R.C., 1991. Mineral requirements of low-birth-weight infants. 
Journal of the American College of Nutrition 10, 474-486. 

Koopman, K.F., 1979. Zoogeography of mammals from islands off the northeastern coast 
of New Guinea. American Museum Novitates 2690, 1-17. 

Koopman, K.F., 1993. Chiroptera. In: Wilson, D.E., Reeder, D.S., (Eds.), Mammal 
Species of the World: A Taxonomic and Geographic Reference. 2 nd ed. 
Smithsonian Institution Press, Washington, D.C. 

Koopman K.F., Steadman, D.W., 1995. Extinction and biogeography of bats on 'Eau, 
Kingdom of Tonga. American Museum Novitates. 3125, 1-3. 

Kozakiewicz, M., 1995. Resource tracking in space and time. In: Hansson, L., Fahrig, L., 
Merriam, G, (Eds.) Mosaic Landscapes and Ecological Processes. Chapman and 
Hall, London. 

Kress, J.W., 1985. Bat pollination of an Old World Heliconia. Biotropica 17, 302-308. 

Kung, A.W.C., Luk, K.D.K., Chu, L.W., Chiu, P.K.Y., 1998. Age-related osteoporosis in 
Chinese: an evaluation of the response of intestinal calcium absorption and 
calcitropic hormones to dietary calcium deprivation. American Journal of Clinical 
Nutrition 68, 1291-1297. 

Kunz, T.H., 1996. Obligate and opportunistic interactions of Old- World tropical bats and 
plants. In: Hasan, Z.A.A., Akbar, Z., (Eds.), Conservation and Faunal 
Biodiversity in Malaysia. Penerbit Universiti Kebangsaan, Malaysia. 

Kunz, T.H., Diaz, C.A., 1995. Folivory in fruit bats with new evidence from Aritbeus 
jamaicensis. Biotropica 27, 106-120. 

Kunz, T.H., Ingalls, K.A., 1994. Folivory in bats: an adaptation derived from frugivory. 
Functional Ecology 8, 665-668. 

Kunz, T.H., Hood, W.R., 2000. Parental care and postnatal growth in Chiroptera. In: 

Crichton, E.G., Krutzsch, P.H., (Eds.), Reproductive Biology of Bats. Academic 
Press, New York. 

Kunz, T.H., Jones, D. P., 2000. Pteropus vampyrus. Mammalian Species 642, 1-6. 

Kunz, T.H., Oftedal, O.T., Robson, S.R., Kretzmann, M.B., Kirk, C, 1995. Changes in 
milk composition during lactation in three species of insectivorous bats. Journal 
of Comparative Physiology- Biology 164, 545-551. 

Kunz, T. H., Stern, A. A., 1995. Maternal investment and post-natal growth in bats. 
Symposium of the Zoological Society of London 67, 123-138. 

Kurta, A., Bell, G.P., Nagy, K.A., Kunz, T.H., 1989. Water balance of free-ranging little 
brown bats (Myotis lucifugus) during pregnancy and lactation. Canadian Journal 
of Zoology 67, 2468-2472. 



131 



Kurta, A., Kunz, T.H., 1987. Size of bats at birth and maternal investment during 
pregnancy. Symposium of the Zoological Society of London 57, 79-106. 

Kwiecinski, G.G., Krook, L., Wimsatt, W.A., 1987a. Annual skeletal changes in the little 
brown bat, Myotis lucifugus lucifugus, with particular reference to pregnancy and 
lactation. The American Journal of Anatomy 178, 410-420. 

Kwiecinski, G.G., Lu, Z.R., Chen, T.C., Holick, M.F., 2001. Observations on serum 25- 
hydroxyvitamin D and calcium concentrations from wild-caught and captive 
neotropical bats, Artibeus jamaicensis. General and Comparative Endocrinology 
122,225-231. 

Kwiecinski, G.G., Wimsatt, W.A., Krook, L, 1987b. Morphology of thyroid C-cells and 
parathyroid glands in summer-active little brown bats, Myotis lucifugus lucifugus, 
with particular reference to pregnancy and lactation. American Journal of 
Anatomy 178, 421-427. 

Law, B.S., 1992. The maintenance nitrogen requirements of the Queensland blossom bat 
(Syconycteris australis) on a sugar/pollen diet: is nitrogen a limiting resource? 
Physiological Zoology 65, 634-648. 

Law, B.S., 1993a. Roosting and foraging ecology of the Queensland blossom 

{Syconycteris australis) in north-eastern New South Wales- flexibility in response 
to seasonal variation. Wildlife Research 20, 419-431. 

Law, B.S., 1993b. Sugar preferences of the Queensland blossom bat, Syconycteris 
australis: a pilot study. Australian Mammalogy 16, 17-21. 

Law, B.S., 1994a. Climatic limitation of the southern distribution of the common 

blossom bat, Syconycteris australis, in New South Wales. Australian Journal of 
Ecology 19, 425-434. 

Law, B.S., 1994b. Banksia nectar and pollen: dietary items affecting the abundance of the 
common blossom bat, Syconycteris australis, in southeastern Australia. Australian 
Journal of Ecology 19, 425-434. 

Law, B.S., 1995. The effect of energy supplementation on the local abundance of the 
common blossom bat, Syconycteris australis, in south-eastern Australia. Oikos 
72, 42-50. 

Law, B.S., 1996. Residency and site fidelity of marked populations of the common 
blossom bat Syconycteris australis in relation to the availability of Banksia 
inflorescences in New South Wales, Australia. Oikos 77, 447-458. 

Law, B.S., 2001. The diet of the common blossom bat (Syconycteris australis) in upland 
tropical rainforest and the importance of riparian areas. Wildlife Research 28 
619-626. 

Law, B.S., Lean, L., 1999. Common blossom bats (Syconycteris australis) as pollinators 
in fragmented Australian tropical rainforest. Biological Conservation 91, 201-212. 



132 

LeBlanc, D., 1999. Bat enrichment guidelines. Enrichment Notebook. 2 nd Ed. American 
Association of Zoo Keepers Topeka, KS. 

Lekagul, B., McNeely, J. A., 1977. Mammals of Thailand. Association for the 
Conservation of Wildlife, Bangkok, Thailand. 

Lemke, T. O., 1984. Foraging ecology of the long-nosed bat, Glossophaga soricina, with 
respect to resource availability. Ecology 65, 538-548. 

Lima, S.L., Zollner, P.A., 1996. Towards a behavioral ecology of ecological landscapes 
Trends in Ecology and Evolution 11, 131-135. 

Loebel, M.R., Sanewski, G., 1987. Flying foxes (Chiroptera: Pteropodidae) as orchard 
pests. Australian Mammalogy 10, 147-150. 

Lowry, J.B., 1989. Green-leaf fractionation by fruit bats: is this feeding behaviour a 

unique nutritional strategy for herbivores? Australian Wildlife Research 16, 203- 
206. 

Marshall, A.G., 1983. Bats, fruits and flowers: an evolutionary relationship in the Old 
World. Biological Journal of the Linnean Society 20, 1 15-135. 

Marshall, A.G., 1985. Old World phytophagous bats (Megachiroptera) and their food 
plants: a survey. Zoological Journal of the Linnean Society 83, 351-369. 

Martinez del Rio, C, 1990. Sugar preferences in hummingbirds: the influence of subtle 
chemical differences on food choice. The Condor 92, 1022-1030. 

Matson, W.J. Jr., 1980. Herbivory in relation to plant nitrogen content. Annual Review of 
Ecological Systematics 11, 119-161. 

Mattingly, D.K., McClure, P. A., 1982. Energetics of reproduction in large-littered cotton 
rats (Sigmodon hispidus). Ecology 63, 183-195. 

McDowell, L.R., 1992. Minerals in Animal and Human Nutrition. Academic Press, New 
York. 

McKean, J.L., 1972. Notes on some collections of bats (Order Chiroptera) from Papua 
New Guinea and Bougainville Island. C.S.I.R.O. Division of Wildlife Research 
Technical Paper 26, 1-35. 

McKean, J.L., 1983. Northern blossom bat. In: Strahan, R., (Ed.), The Australian 

Museum Complete Book of Australian Mammals. Angus and Robertson, Sydney. 

McNab, B.K., Bonaccorso, F.J., 1995. The energetics of pteropodid bats. Symposium of 
the Zoological Society of London 67, 1 1 1-122. 

McNab, B.K., Bonaccorso, F.J., 2001. The metabolism of New Guinean Pteropodid bats. 
The Journal of Comparative Physiology B- Biochemical, Systematic and 
Environmental Physiology 171, 201-214. 

Medway, Lord., 1969. Wild mammals of Malaya (Peninsular Malaysia) and Singapore. 
Oxford University Press, Kuala Lumpur. 

Mickleburgh, S.P., Hutson, A.M., Racey, P. A., 1992. Old World Fruit Bats: An Action 
Plan for Their Conservation. RJCN, Switzerland. 



133 

Mildenstein, T.L, 2002. Habitat selection of large flying foxes using radio telemetry: 

targeting conservation efforts in Subic Bay, Philippines. MS thesis, University of 
Montana, Missoula. 

Miles, P.H., Wilkinson, N.S., McDowell, L.R., 2001. Analysis of Minerals for Animal 
Nutrition Research. 3 rd Ed. Department of Animal Sciences, The University of 
Florida, Gainesville. 

Millar, J.S., 1977. Adaptive features of mammalian reproduction. Evolution 31, 370-386. 

Miller, C.A., Wilson, D.E., 1997. Pteropus tonganus. Mammalian Species 552, 1-6. 

Miller, E.R., 1980. Bioavailability of minerals. Proceedings of the Minnesota Nutrition 
Conference. University of Minnesota Press, St. Paul. 

Mittermeier, R.A., Goettsch-Mittermeier, C, Robles Gil, P., 1997. Megadiversity: 
Earth's Biologically Wealthiest Nations. Cemex, Mexico. 

Moe, S.R., 1973. Mineral content and wildlife use of soil licks in southwestern Nepal. 
Canadian Journal of Zoology 71, 933-936. 

Mohd-Azlan, J., Zubaid, A., Kunz, T.H., 2001. Distribution, relative abundance, and 
conservation status of the large flying fox, Pteropus vampyrus, in peninsular 
Malaysia: a preliminary assessment. Acta Chiropterologica 3, 149-162. 

Montenegro, O.L., 1998. The behavior of lowland tapir (Tapirus terrestris) at a natural 
mineral lick in the Peruvian Amazon. M.S. thesis, University of Florida, 
Gainesville. 

Mudar, K.M., Allen, M.S., 1986. A list of bats from northern Luzon, 
Philippines.Mammalia 50, 219-225. 

Muller, L.D., Schaffer, L.V., Ham, L.C., Owens, M.J., 1977. Cafeteria style free-choice 
mineral feeder for lactating dairy cows. Journal of Dairy Science 60, 1574-1582. 

Nams, V.O., 2001. Locate 2, User's Manual. Nova Scotia, Canada. 

National Oceanic and Atmospheric Administration (NOAA), 1996. Local 

climatologicaldata: annual summary with comparative data. Pago Pago, American 
Samoa.National Oceanic and Atmospheric Administration, National Climatic 
Data Center Report, Denver. 

National Oceanic and Atmospheric Administration (NOAA). 1999. Local climatological 
data: annual summary with comparative data. Pago Pago, American Samoa 
National Oceanic and Atmospheric Administration, National Climatic Data 
Center Report, Denver. 

National Research Council (NRC), 1989. Recommended Daily Allowances. 10 th ed., 
National Academy Press, Washington, D.C. 

National Research Council (NRC), 1995. Nutrient requirements of laboratory animals. 4 th 
ed. National Academy Press, Washington, D.C. 

National Research Council (NRC), 2001 . Nutrient Requirements of Dairy Cattle. 7 th ed. 
National Academy Press, Washington D.C. 



134 



Nelson, J.E., 1965. Movements of Australian flying foxes (Pteropodidae: 
Megachiroptera). Australian Journal of Zoology 13, 53-73. 

Nelson, S.L., Miller, M.A., Heske, E.J., Fahey, G.C. Jr., 2000a. Nutritional consequences 
of a change in diet from native to agricultural fruits for the Samoan fruit bat. 
Ecography 23, 393-401. 

Nelson, S.L., Miller, M.A., Heske, E.J., Fahey, G.C. Jr., 2000b. Nutritional quality of 
leaves and unripe fruit consumed as famine foods by the flying foxes of Samoa. 
Pacific Science 54, 301-311. 

Neu, C.W., Byers, C.R., Peek, J.M., 1974. A technique for analysis of utilization- 
availability data. Journal of Wildlife Management 38, 541-545. 

Nicolay, C.W., Dumont, E.R., 2000. An experimental analysis of feeding performance in 
Syconycteris australis (Megachiroptera, Pteropodidae). Mammalia 64, 155-161. 

Norusis, M. J., 1993. SPSS for Windows. Professional Statistics, Release 6.0. SPSS, 
Chicago, IL. 

Nowak, R.M. 1999. Walker's Mammals of the World. 6 th ed. The Johns Hopkins 
University Press, Baltimore. 

O'Brien, T.G., Kinnard, M.F., Dierenfeld, E.S., Conklin, N.L., Wrangham, R.W., Silver, 
S.C., 1998. What's so special about figs? Nature 292, 668. 

Oftedal, O.T., 1991. The nutritional consequences of foraging in primates: the 
relationship of nutrient intake to nutrient requirements. Philosophical 
Transactions of the Royal Society of London B 334, 161-170. 

Oftedal, O.T., Allen, M.E., 1996. The feeding and nutrition of omnivores with emphasis 
on primates. In: Kleiman, D.G., Allen, M.E., Thompson, K.V., Lumpkin, S., 
(Eds.), Wild Animals in Captivity; Principles and Techniques. The University of 
Chicago Press, Chicago. 

Okon, E.E., 1977. Functional anatomy of the alimentary canal in the fruit bat Eidolon 
helvum and the insect bat Tadarida nigeriae. Acta Zoologica 58, 83-93. 

Ozanne, P.G., Howes, K.M.W., 1971. Effects of grazing on phosphorous requirement of 
an annual pasture. Australian Journal of Agricultural Research 22, 81-85. 

Palmer, C, Price, O., Bach, C, 2000. Foraging ecology of the black flying fox {Pteropus 
alecto) in the seasonal tropics of the Northern Territory, Australia. Wildlife 
Research 27, 169-178. 

Palmer, C, Woinarski, J. C. Z., 1999. Seasonal roosts and foraging movements of the 

black flying fox {Pteropus alecto) in the Northern Territory: resource tracking in a 
landscape mosaic. Wildlife Research 26, 823-838. 

Pansu, D., Duflos, C, Bellaton, C, Bronner, F., 1993. Solubility and intestinal transit- 
time limit calcium absorption in rats. Journal of Nutrition 123, 1396-1404. 

Papadimitriou, H.M., Swartz, S.M., Kunz, T.H., 1996. Ontogenic and anatomic variation 
in mineralization of the wing skeleton of the Mexican free-tailed bat, Tadarida 
brasiliensis. Journal of Zoology (London) 240, 41 1-426. 



135 



Parry-Jones, K., Augee, M.L., 1991a. Food selection by grey-headed flying foxes 

{Pteropus poliocephalus) occupying a summer colony site near Gosford, New 
South Wales. Wildlife Research 18, 1 1 1-124. 

Payne, J., Francis, CM., Phillips, K., 1985. A Field Guide to the Mammals of Borneo. 
The Sabah Society, Kota Kinabulu, Malaysia. 

Peale, T.R., 1848. Zoology. United States exploring expedition. Vol. 8. C. Sherman. 
Philadelphia. 

Perkin-Elmer Corporation, 1980. Analytical methods for atomic absorption 
spectrophotometry. Perkin-Elmer, Norwalk. 

Petchey, F.J., 2001. Radiocarbon determinations from the Mulifauna Lapita site, Upolu, 
Western Samoa. Radiocarbon 43, 63-68. 

Pettigrew, J.D., 1991 . Wings or brain- convergent evolution in the origins of bats. 
Systematic Zoology 40, 199-216. 

Pierson, E.D., Elmqvist, T., Rainey, W.E., Cox, P. A., 1996. Effects of tropical cyclonic 
storms on flying foxes on the South Pacific Islands of Samoa. Conservation 
Biology 10, 438-451. 

Pierson, E.D., Rainey, W.E., 1992. The biology of flying foxes in the genus Pteropus: 
review. In: Wilson, D.E., Graham, G.L. (Eds.), Pacific Island Flying Foxes 
Proceedings of an International Conservation Conference. U.S. Fish and Wildlife 
Service., Washington, D.C. 

Provenza, F.D., Scott, C.B., Phy, T.S., Lynch, J. J., 1996. Preference of sheep for foods 
varying in flavors and nutrients. Journal of Animal Science 74, 2355-2361. 

Quoy J.R.C., Gaimard, J.P., 1830. Voyages de decouvertes del'Astrolabe execute 
parordre du roi, pendant les annees 1826, 1827,1828, 1829 sous le 
commandement de M.J. Dumont d'Urville. Zoologie 1, 74-77. 

Racey, P. A., 1979. Two bats in the Seychelles. Oryx 15, 148-152. 

Racey, P. A., 1988. Reproductive assessment in bats. In: Kunz, T.H., (Ed.), Ecological 
and Behavioral Methods for the Study of Bats. Smithsonian Institution Press, 
Washington, D.C. 

Racey, P.A., Entwistle, A.C., 2000. Life-history and reproductive strategies of bats. In: 
Crichton, E.G., Krutzsch, P.H., (Eds.), Reproductive Biology of Bats. Academic 
Press, New York. 

Racey, P.A., Speakman, J. R., 1987. The energy costs of pregnancy and lactation in 

heterothermic bats. Symposium of the Zoological Society of London 57, 107-125. 

Radostits, O.M., Blood, D.C, Gay, C.C, 1994. Veterinary Medicine: a Textbook of the 
Diseases of Cattle, Pigs, Goats and Horses. 8 th ed. Bailliere Tindall, Philadelphia. 

Rainey, W.E., 1990. The flying fox trade: becoming a rare commodity. Bats 8, 6-9. 

Rainey, W.E., 1998. Conservation of bats in remote Indo-Pacific Islands. In: Kunz, T.H., 
Racey, P.A., (Eds.), Bat Biology and Conservation. Smithsonian Institution Press, 
Washington, D.C. 



136 

Rainey, W.E., Pierson, E.D., Elmqvist, T., Cox, P.A., 1995. The role of flying foxes 

(Pteropodidae) in oceanic island ecosystems of the Pacific. In: Racy, P. A., Swift, 
S.M. (Eds.), Ecology, Evolution and Behavior of Bats. Clarendon Press, Oxford. 

Rajan, K.M., Nair, N.G., Subbaraj, R., 1999. Seasonal food preference of the Indian 
short-nosed fruit bat Cynopterus sphinx (Vahl) (Chiroptera: Pteropodidae). 
Journal of the Bombay Natural History Society 9, 24-27. 

Recommended Dietary Allowances (RDA), 1989. 10 th ed. National Academy Press. 
Washington D.C. 

Richards, G.C., 1983. Queensland blossom-bat. In: Strahan, R., (Ed.), The 

AustralianMuseum Complete Book of Australian mammals. Angus and 
Robertson, Sydney 

Richards, G.C., 1990. The spectacled flying fox, Pteropus conspicullatus, (Chiroptera: 
Pteropodidae) in north Queensland. 1 . Roost sites and distribution patterns. 
Australian Mammalogy 13, 17-24. 

Robbins, C.T., 1993. Wildlife Feeding and Nutrition. Academic Press, New York. 

Roces, F., Winter, Y., von Helverson, O., 1993. Nectar concentration preference and 
water balance in a flower visiting bat, Glossophaga soricina antillarum. In: 
Barthlott, W. (Ed.), Plant-Animal Interactions in Tropical Environments. Museum 
of Alexander Koenig, Bonn. 

Ruby, J., Nathan, P.T., Balasingh, J., Kunz, T.H., 2000. Chemical composition of fruits 
and leaves eaten by the short-nosed fruit bat, Cynopterus sphinx. Journal of 
Chemical Ecology 26, 2825-2841. 

SAS, 1999. SAS/Stat User's Guide, Version 8. SAS Institute, Inc., Cary, N.C. 

Senft, R.L., Coughenour, M.B., Bailey, D.W., Rittenhouse, L.R., Sala, O.E., Swift, D.M., 
1987. Large herbivore foraging and ecological hierarchies. Bioscience 37, 789. 

Sevick, S.H., Studier, E.H., 1992. Change in forearm curvature throughout the summer in 
female big brown bats. Journal of Mammalogy 73, 866-870. 

Shilton, LA., Altingham, J.D., Compton, S.G., and Whittacker R.J., 1999. Old 

Worldfruit bats can be long-distance seed dispersers through extended retention of 
viable seeds in the gut. Proceedings of the Royal Society, Biological Sciences 
266,219-223. 

Soares, J.H. Jr., 1995. Calcium bioavailability. In: Ammerman, C.B., Baker, D.H., Lewis, 
A. J., (Eds.), Bioavailability of Nutrients for Animals: Amino Acids, Minerals and 
Vitamins. Academic Press, New York. 

Sokal, R.R., Rohlf, F.J., 1995. Biometry. W.H. Freeman and Company, San Francisco. 

Spencer, H.J., Fleming, T.H., 1989. Roosting and foraging behavior of the Queensland 
tube-nosed bat, Nyctimene robinsoni (Pteropodidae): preliminary radio-tracking 
observations. Australian Wildlife Research 16, 413-420. 

Spencer, H.J., Palmer, C, Parry-Jones, K., 1991. Movements of fruit-bats in Eastern 
Australia, determined by using radio-tracking. Wildlife Research 18, 463-468. 



137 



Springer, J.T., 1979. Some sources of bias and sampling error in radio triangulation. 
Journal of Wildlife Management 43, 926-935. 

Start, A.N., Marshall, A.G., 1974. Nectivorous bats as pollinators of trees in West 
Malaysia. In: Burley, J., Stiles, B.T., (Eds.), Variation, Breeding, and 
Conservation of Forest Trees. Academic Press, London. 

Steadman, D.W., 1991. Extinct and extirpated birds from Aitutaki and Atiu, southern 
Cook Islands. Pacific Science 45, 325-347. 

Steadman, D.W., Kirch, P.V., 1990. Prehistoric extinctions of birds on Mangaia, Cook 
Islands, Polynesia. Proceedings of the National Academy of Science 87, 9605- 
9609. 

Steller, D.C., 1986. The dietary energy and nitrogen requirements of the grey-headed 
flying fox, Pteropus poliocephalus (Temminck) (Megachiroptera). Australian 
Journal of Zoology 34, 339-349. 

Stier, S.C., Mildenstein, T.L., 2001. A dietary investigation of co-roosting endemic and 
endangered large flying foxes in the Philippines. 15 th Meeting of the Society for 
Conservation Biology. Hilo, Hawaii. July 29- August 1, 2001. 

Strickland, G.T., Beckner, W.M., Leu, M.L., 1972. Absorption of copper in homozygotes 
and heterozygotes for Wilson's disease and controls- isotope tracer studies with 
Cu-67 and Cu-64. Clinical Science 43, 617. 

Studier, E.H., Kunz, T.H., 1995. Accretion of nitrogen and minerals in suckling bats, 
Myotis velifer and Tadarida brasiliensis. Journal of Mammalogy 76, 32-42. 

Studier, E.H., Sevick, S.H., Keeler, J.O., Schenck, R.A., 1994a. Nutrient levels in guano 
from maternity colonies of big brown bats. Journal of Mammalogy 75, 71-83. 

Studier, E.H., Sevick, S.H., Ridley, D.M., Wilson, D.E., 1994b. Mineral and nitrogen 

concentrations in feces of some neotropical bats. Journal of Mammalogy 75 674- 
680. 

Studier, E.H., Viele, D.P., Sevick, S.H., 1991. Nutritional implications for nitrogen and 
mineral budgets from analysis of guano of the big brown bat Eptesicus fuscus 
(Chiroptera, Vespertilionidae). Comparative Biochemistry and Physiology A- 
Physiology 100, 1035-1039. 

Suttle, N.F., 1991. The interactions between copper, molybdenum and sulphur in 
ruminant nutrition. Annual Review of Nutrition 1 1, 121-140. 

Tan, K.H., Zubaid, A., Kunz, T.H. 1998. Food habits of Cynopterus brachyotis (Muller) 
(Chiroptera: Pteropodidae) in Peninsular Malaysia. Journal of Tropical Ecology 
14, 299-307. 

Tedman, R.A., Hall, L.S., 1985. The morphology of the gastrointestinal tract and food 
transit time in the fruit bats Pteropus alecto and Pteropus poliocephalus. 
(Megachiroptera) Australian Journal of Zoology 33, 625- 640. 



138 



Thornton, I.W.B., Compton, S.G., Wilson, C.N., 1996. The role of animals in the 
colonization of Krakatau Island by fig trees (Ficus species). The Journal of 
Biogeography 23, 577-592. 

Thomas, D.W., 1984. Fruit intake and energy budgets of frugivorous bats. Physiological 
Zoology 57, 457-457. 

Thompson, S.D. 1991. Energetics of gestation and lactation in small mammals: basal 
metabolic rate and the limits of energy use. In: Tomasi, T., Horton, T. (Eds.), 
Mammalian energetics, Interdisciplinary Views of Metabolism and Reproduction. 
Cornell University Press, New York. 

Thomson, S.C., Brooke, A.P., Speakman, J.R., 1998. Diurnal activity in the Samoan 

flying fox, Pteropus samoensis. Philosophical Transactions of the Royal Society 
of London, Series B 353, 1595-1606. 

Tidehag, P., Moberg, A., Sunzel, B., Hallmans, G., Sjostrom, R., Wing, K., 1988. The 
availability of zinc, cadmium, and iron from different grains measured as isotope 
absorption and mineral accumulation in rats. In: Hurley, L.S., Keen, C.L., 
Lonnerdal, B., Rucker, R.B., (Eds.), Trace Elements in Man and Animals. Plenum 
Press, New York. 

Tracy, B.F., McNaughton, S.J., 1995. Elemental analysis of mineral lick soils from the 
Serengeti National Park, the Konza Prairie, and Yellowstone National Park. 
Ecography 18, 91-94. 

Trail, P., 1994. The phenology of rainforest plants in Tutuila, American Samoa. 
Department of Marine and Wildlife Resources, American Samoa. 

Uhland, A.M., Kwiecinski, G.G., DeLuca, H.F., 1992. Normalization of serum-calcium 
restores fertility in Vitamin-D-deficient male rats. Journal of Nutrition 122 1338- 
1344. 

Walinski, I., Guggenheim, K., 1974. Effect of low calcium diet on bone and calcium 
metabolism in rats and mice- a differential species response. Comparative 
Biochemistry and Physiology 49 A, 183-195. 

Wasserman, R.H., 1981. Intestinal absorption of calcium and phosphorus. Federation 
Proceedings 40, 68-72. 

West, C.C., 1986. Reproductive biology of Pteropus rodricensis. Myotis 23, 137-141. 

Whistler, W.A., 1992. Vegetation of Samoa and Tonga. Pacific Science 46, 159-178. 

Whistler, W.A., 1994. Botanical inventory of the proposed Tutuila and Ofu units of the 
National Park of American Samoa. University of Hawaii Cooperative National 
Park Resources Studies Tech.Report 87, University of Hawai'i, Manoa 

Whittacker, R.J., Jones, S.H., 1994. The role of frugivorous bats and birds in the 
rebuilding of a tropical forest ecosystem, Krakatau, Indonesia. Journal of 
Biogeography 21, 245-258. 






139 



Wickler, W., Seibt, U., 1964. Field studies of the African fruit bat, Epomophorus 

wahlbergi (Sundovall), with special reference to male calling. Zeitschrift Fur 
Tierpsychologie 40, 345-376. 

Wiles, G.J., 1987b. Current research and future management of Marianus fruit bats 
(Chiroptera: Pteropodidae) on Guam. Australian Mammalogy 10, 93-95. 

Wiles, G.J., Engbring, J., Otobed, D., 1997. Abundance, biology, and human exploitation 
of bats in the Palau Islands. Journal of Zoology, London 241, 203-227. 

Wiles, G.J., Fujita, M.S., 1992. Food plants and economic importance of flying foxes on 
Pacific Islands. In: Wilson, D.E., Graham, G.L. (Eds.), Pacific Island Flying 
Foxes: Proceedings of an International Conservation Conference. U.S. Fish and 
Wildlife Service, Washington D.C. 

Wiles, G.J., Payne, N.H., 1986. The trade in fruit bats, Pteropus spp. on Guan and other 
Pacific islands. Biological Conservation 38, 143-161. 

Wilson, D.E., 1988. Maintaining bats for captive studies. In: Kunz, T.H., (Ed.), 

Ecological and Behavioral Methods for the Study of Bats. Smithsonian Institution 
Press, Washington, D.C. 

Wilson, D.E., Engbring, J., 1992. The flying foxes Pteropus samoensis and Pteropus 

tonganus: status in Fiji and Samoa. In: Wilson, D.E., Graham, G.L. (Eds.), Pacific 
Island Flying Foxes: Proceedings of an International Conservation Conference. 
U.S. Fish and Wildlife Service, Washington D.C. 

Winkelmann, J.R., Bonaccorso, F.J., Strickler, T.L., 2000. Home range of the southern 
blossom bat, Syconycteris australis, in Papua New Guinea. Journal of 
Mammalogy 81, 408-414. 

Witmer, M.C., 1998. Ecological and evolutionary implications of energy and protein 
requirements of avian frugivores eating sugary diets. Physiological Ecology 71, 
599-610. 

Wodzicki, K., Felton, H., 1980. Fruit bats of the genus Pteropus from the islands 

Rarotonga and Mangaia, Cook Islands, Pacific Ocean (Mammalia: Chiroptera). 
Senckenbergianan Biologica 61, 43-151. 

Zervas, G., Rissaki, M., Deligeorgis, S., 2001. Free-choice consumption of mineral lick 
blocks by fattening lambs fed ad libitum alfalfa hay and concentrates with 
different trace mineral content. Livestock Production Science 68, 251-258. 

Zimmerman, J.W., Powell, R.A., 1995. Radiotelemetry error- location error method 
compared with error polygons and confidence ellipses. Canadian Journal of 
Zoology 73, 1123-1133. 

Zollner, P. A., Lima, S.L., 1999. Search strategies for landscape-level interpatch 
movements. Ecology 80, 1019-1030. 

Zortea, M., Mendes, S.L., 1993. Folivory in the big fruit-eating bat, Artibeus lituratus 
(Chiroptera: Phyllostomidae) in eastern Brazil. Journal of Tropical Ecology 9, 1 17-120. 



BIOGRAPHICAL SKETCH 
Suzanne Nelson was born on April 12, 1972. She was raised in Mt. Prospect, 
Illinois, a suburb northwest of Chicago. She was the middle of three children. As a child 
she loved to be outside and spent most of her vacations at her grandparent's house in 
south Florida and at her aunt and uncle's farm in southern Illinois. After graduating from 
Prospect High School in 1990, she attended the University of Illinois at Urbana- 
Champaign and graduated in 1994 with a B.S. degree in Ecology, Ethology, and 
Evolution. She traveled to American Samoa for the first time as a volunteer wildlife 
biologist while taking a year off to travel after college. In 1995 she returned to the 
University of Illinois to study nutritional ecology of fruit bats in American Samoa for her 
M.S. degree with Dr. Edward Heske. She graduated with an M.S. in Biology in 1997. She 
taught at the University of Alaska- Anchorage before beginning her Ph.D. program at the 
University of Florida at Gainesville in 1998. While at the University of Florida, she was 
supported by the Luis F. Bacardi Graduate Fellowship as a student of Dr Steve 
Humphrey and Dr. Tom Kunz. Suzanne lives with her husband Darrin Masters and their 
golden retriever, Louie. 



140 



I certify that I have read this study and that in my opinion it conforms to 
acceptable standards of scholarly presentation and is fully adequate, in scope and quality, 
as a dissertation for the degree of Doctor of Philosophy. 




StepWen R. Humphrey, (rhair 
Professor of Wildlife Ecology 
and Conservation 

I certify that I have read this study and that in my opinion it conforms to 
acceptable standards of scholarly presentation and is fully adequate, in scope and quality, 
as a dissertation for the degree of Doctor of Philosot 




Thomas 
Profes; 

and Conservation 



I certify that I have read this study and that in my opinion it conforms to 
acceptable standards of scholarly presentation and is fully adequate, imscope and quality, 
as a dissertation for the degree of Doctor of Philosophy 




lelvin E. Sunquist 
Associate Professor of Wildlife Ecology 
and Conservation 

I certify that I have read this study and that in my opinion it conforms to 
acceptable standards of scholarly presentation and is fully adequate, in scope and quality, 
as a dissertation for the degree of Doctor of Philosophy. 




George W: Tanner 
Professor of Wildlife Ecology and 
Conservation 

I certify that I have read this study and that in my opinion it conforms to 
acceptable standards of scholarly presentation and is fully adequate, in scope and quality, 
as a dissertation for the degree of Doctor of Philosophy. 

Lee R. McDowell 
Professor of Animal Sciences 



This dissertation was submitted to the Graduate Faculty of the College of 
Agricultural and Life Sciences and to the Graduate School and was accepted as partial 
fulfillment of the requirements for the degree of Doctor of Philosophy. 



May 2003 




V J^W- 




Dean, College of Agricultural and L^fe 
Sciences 




Dean, Graduate School 































uw 



UNIVERSITY OF FLORIDA 

lllllllilllllK 

3 1262 08557 1833